Episodic Behavior of the Jordan Valley Section of the Dead Sea
Fault Inferred from a 14-ka-Long Integrated
Catalog of Large Earthquakes
by Matthieu Ferry,* Mustapha Meghraoui, Najib Abou Karaki,Masdouq Al-Taj, and Lutfi Khalil
Abstract The continuous record of large surface-rupturing earthquakes along theDead Sea fault brings unprecedented insights for paleoseismic and archaeoseismicresearch. In most recent studies, paleoseismic trenching documents the late Holocenefaulting activity, while tectonic geomorphology addresses the long-term behavior(>10 ka), with a tendency to smooth the effect of individual earthquake ruptureevents (Mw >7). Here, we combine historical, archaeological, and paleoseismicinvestigations to build a consolidated catalog of destructive surface-rupturing earth-quakes for the last 14 ka along the left-lateral Jordan Valley fault segment. The 120-km-long fault segment limited to the north and the south by major pull-apart basins(the Hula and the Dead Sea, respectively) is mapped in detail and shows five subseg-ments with narrow stepovers (width < 3 km). We conducted quantitative geomor-phology along the fault, measured more than 20 offset drainages, excavated fourtrenches at two sites, and investigated archaeological sites with seismic damage inthe Jordan Valley. Our results in paleoseismic trenching with 28 radiocarbon datingsand the archaeoseismology at Tell Saydiyeh, supplemented with a rich historical seis-mic record, document 12 surface-rupturing events along the fault segment with a meaninterval of ∼1160 yr and an average 5 mm=yr slip rate for the last 25 ka. The mostcomplete part of the catalog indicates recurrence intervals that vary from 280 yr to1500 yr, with a median value of 790 yr, and suggests an episodic behavior for theJordan Valley fault. Our study allows a better constraint of the seismic cycle and re-lated short-term variations (late Holocene) versus long-term behavior (Holocene andlate Pleistocene) of a major continental transform fault.
Introduction
The occurrence of large earthquakes on continentalfaults holds crucial questions on their physical and mechan-ical characteristics, their size, and their time distribution interms of magnitude and frequency. Recent field investiga-tions in paleoseismology and archaeoseismology along theDead Sea fault (DSF) show evidence of historical coseismicsurface rupturing at the Sicantarla Tell in Turkey (Amikbasin; Altunel et al., 2009), the Al-Harif Roman aqueductin Syria (Meghraoui et al., 2003), the Lebanese restrainingbend (Gomez et al., 2003; Daëron et al., 2004; Nemer andMeghraoui, 2006; Daëron et al., 2007; Nemer et al., 2008),the Jordan Valley gorge and Hula basin (Ellenblum et al.,1998; Marco et al., 2003; Marco et al., 2005), the JordanValley (Reches and Hoexter, 1981), and the Wadi Araba (Zil-
berman et al., 2005; Haynes et al., 2006). Paleoseismicstudies along the Jordan Valley fault (JVF) revealed theepisodic activity from a long-term earthquake record(50 ka) in the Lisan lacustrine deposits (El-Isa and Mustafa,1986; Marco et al., 1996; Migowski et al., 2004) and fromthe correlation between cumulative stream offsets and 48-ka-long paleoclimatic fluctuations (Ferry et al., 2007). This epi-sodic activity is expressed not only by periods of earthquakeclusters affecting a single segment but also by a sequence ofearthquakes on different segments during a short period oftime (Ambraseys, 2004; Sbeinati et al., 2005). The long-termrecord of past earthquakes contributes to better understandthe faulting behavior and stability of segment boundaries(Sieh, 1996), as well as fault interactions during earthquakesequences (Stein et al., 1997). Although earthquake-inducedsoft-sediment deformations were largely studied in the Lisanformation, earthquake surface ruptures associated with the*Also at Institut de Physique du Globe, Strasbourg, France.
39
Bulletin of the Seismological Society of America, Vol. 101, No. 1, pp. 39–67, February 2011, doi: 10.1785/0120100097
JVF needed a detailed paleoseismic and archaeoseismicstudy in order to document the earthquake sequence andrelated seismic cycle on a single fault segment.
The DSF forms the boundary between the African andArabian plates and accommodates ∼1 cm=yr of relative left-lateral strike-slip motion (Quennell, 1981; Fig. 1). The fault
system exhibits a relatively simple geometry with largepull-apart basins distributed along strike (the Red Sea, theDead Sea, the Hula basin, the Ghab basin, and the Amikbasin) and a single major restraining bend at its center(Mounts Lebanon and Anti-Lebanon). It is composed ofeight major segments (Fig. 1a), all of which are capable
Figure 1. (a) General map of the Dead Sea Transform system. Numbers are geological slip rates (in black) and geodetic strain rates (inwhite). Sources: Klinger et al. (2000); Niemi et al. (2001); Meghraoui et al. (2003); Reilinger et al. (2006); Ferry et al. (2007). Pull-apartbasins: ab, Amik basin; gb, Ghab basin; hb, Hula basin; ds, Dead Sea. Major fault segments: EAF, East Anatolian fault; AF, Afrin fault; KF,Karasu fault; JSF, Jisr Shuggur fault; MF, Missyaf fault; YF, Yammouneh fault; ROF, Roum fault; RAF, Rachaya fault; SF, Serghaya fault;JVF, Jordan Valley fault; WAF, Wadi Araba fault. (b) Detailed map of the JVF segment between the Sea of Galilee and the Dead Sea. Thesegment itself is organized as six 15-km to 30-km-long right-stepping subsegments limited by 2-km to 3-km-wide transpressive relay zones.The active trace of the JVF continues for a further ∼10 km northward into the Sea of Galilee (SG) and ∼20 km southward into the northernDead Sea (DS). The color version of this figure is available only in the electronic edition.
40 M. Ferry, M. Meghraoui, N. Abou Karaki, M. Al-Taj, and L. Khalil
of producing large destructive earthquakes and surface fault-ing as documented by an extended historical record (e.g.,Guidoboni et al., 1994; Ambraseys and Jackson, 1998; Sbei-nati et al., 2005; Ambraseys, 2009). However, other than the1995 Mw 7.3 Aqaba earthquake, none of the main segmentshas released a large earthquake in the last eight to nine cen-turies (Fig. 2). The lack of seismicity and elapsed time sincethe most recent historical earthquakes suggest that a tectonicloading has been accumulating along most of the DSF.Global Positioning System plate velocities along the faultsystem suggest 2:5–6 mm=yr (Fig. 1a; also, McClusky et al.,2003; Wdowinski et al., 2004; Reilinger et al., 2006; Gomezet al., 2007; Le Beon et al., 2008; Alchalbi et al., 2010) thatare comparable to 4–7 mm=yr geological slip rates (Garfun-kel et al., 1981; Ginat et al., 1998; Klinger et al., 2000;Niemi et al., 2001; Meghraoui et al., 2003; Gomez et al.,2003; Daëron et al., 2004; Akyüz et al., 2006; Ferry et al.,2007; Karabacak et al., 2010; Sbeinati et al., 2010) measuredat 2-to-100-ka time scales. It implies 3–5 m of slipdeficit for the different segments and suggests an increasingpotential for destructive events in the near future.
After a geology, tectonic geomorphology, and seismicitysetting, we present the paleoseismic investigations with fourtrenches across the fault and the archaeoseismic studies at 11different sites, with a specific focus on the Tell Saydiyehda-mages. Our results yield an integrated catalog of faultingevents and related large earthquakes for the last 14 ka. Theearthquake catalog of faulting events that includes both clus-tering and quiescence periods sheds light on the distributionof interseismic periods and the associated tectonic-loadingprocess. The long-term faulting behavior of the JVF andits relationship to neighboring segments is also discussed.
Geological Setting
The north–south-trending DSF transform (Fig. 1) ismade of a transtensional system to the south (including theHula, Dead Sea, and Gulf of Aqaba pull-apart basins), theLebanese restraining bend (the Yammouneh, Rachaya,Seghaya, and Roum faults) in the middle, and a strike-slipsystem to the north (the Missyaf fault and the Ghab pull-apartbasin). The seismicity described by Aldersons et al. (2003)suggests that the lower crust has a brittle behavior below20 km and possibly as deep as 32 km. This is supported bythermomechanical modeling (Petrunin and Sobolev, 2006),which suggests the brittle part of the cold lithosphere beneaththe Dead Sea basin may be locally as thick as 27 km.Similarly, heat-flow measurements indicate relatively lowvalues not typical of a rifting region (Ben-Avraham et al.,1978). Furthermore, Ryberg et al. (2007) image subverticalmajor faults and deep sedimentary basins along the WadiAraba segment. Although the level of background seismicityis low (M <5), the seismotectonic characteristics along theplate boundary show a systematic pattern of strike-slip focalmechanisms and some normal faulting solutions (Salamonet al., 2003). The DSF is a typical continental transform fault
Figure 2. Seismicity of the Dead Sea Transform system.Instrumental events with M ≥4 from 1964 to 2006 (IRIS DataManagement Center; see Data and Resources section) in filledcircles. Background seismicity is very scarce and mainly restrictedto the Lebanese Bend and the Jordan Valley. The 1995 Mw 7.3Aqaba earthquake and aftershock swarm dominate the seismicityof the Red Sea basin. Historical events with I0 ≥ VII (Ambraseysand Jackson, 1998; Sbeinati et al., 2005) in open circles. Apart fromthe 1927 Mw 6.2 Jericho earthquake, no significant event hasoccurred along the JVF since A.D. 1033 (see text for details).The color version of this figure is available only in the electronicedition.
Episodic Behavior of the Jordan Valley Section of the Dead Sea Fault 41
and exhibits a narrow deformation zone dominated by strike-slip faulting that affects a thick cold crust. The post-Mioceneleft-lateral offset along the fault is estimated to be ∼45 km,which is consistent with the accumulation of more than 8 kmof clastic, carbonatic, and evaporitic sediments in the DeadSea basin (Quennel, 1984; ten Brink, 1993; Ginzburg andBen-Avraham, 1997; Bartov et al., 2006). At a large scale,the DSF is highly segmented, and the JVF section is limitedby the Lebanese restraining bend to the north and the largeDead Sea pull-apart basins to the south.
Tectonic Geomorphology alongthe Jordan Valley Fault
The active JVF is made of five 15-to-30-km-long subseg-ments (Fig. 1b; Al-Taj, 2000; Malkawi and Alawneh, 2000;Ferry et al., 2007) limited by relatively small (2-to-3-km-wide) transpressive and transtensive relay zones. Using aerial(1:25000 scale) and satellite photographs (SPOT-5, Landsat7, Google Earth), field investigations, and offset measure-ments, we mapped in detail a total of 120 km of the activefault trace from the Hula basin to the Dead Sea. The activefault trace is visible within the valley and cuts through theformer Lake Lisan (65–18 ka B.P.) and clastic Damya(18 ka B.P. to present) deposits. The geomorphology of thevalley shows regionally flat Lisan lacustrine terraces display-ing an average slope of ∼0:1° toward the present-day DeadSea. The very fine-grained Lisan sediments (mostly varve-like detritus and aragonite), combined with a semiaridclimate produce classical badland morphology, where mate-rial is eroded away through a dense dentritic gully network.The hydrographic system provides clear markers for thestudy of left-lateral cumulative offsets along the fault trace(Ferry et al., 2007; Ferry andMeghraoui, 2008; Klein, 2008).
The JVF displays complex transtensional features withnumerous 100-to-300-m-long and 50-m-wide pull-apartbasins (Fig. 3) separated by right-stepping en echelon ruptures(fig. 3 in Ferry et al., 2007). Located in the southern section ofthe JVF, the impressive Ghor Katar badland area is made ofnumerous stream incisions that expose ∼50-m-thick Damyaand Lisan lacustrine units in remarkable cliffs (Abed andYaghan, 2000). The fault is well visible in the many incisionsof Ghor Katar before it enters the flat lacustrine terrace.
Located 2.5 km south of Ghor Katar, the Ghor Kabedpull-apart system (Fig. 3c) affects the Damya and Lisanterraces and illustrates the pattern of active faulting in thevalley. The accumulation of late Pleistocene and Holocenedeposits in the Ghor Kabed pull-apart depressions constitutesa good record of past faulting events. Indeed, a detailedmicrotopographic survey (1–2-m resolution) illustrates thebasin morphology and related north–south-trending 6-m-high fault scarps with 4–15° slope cutting through the middleof the depocenter. The clear fault scarp morphology andactive alluvial and lacustrine sedimentary processes presenta good potential for paleoseismic trenching at this site (seethe Paleoseismology section).
In the middle section of the JVF, at the Tell Saidiyeh area,small stream offsets and abandoned beheaded channelsprovide an ideal site for faulting event characterization andslip rate calculations (Fig. 3d). We performed a detailedmicrotopographic survey (Fig. 3d) of the site in order to studythe cumulative offsets along the fault revealed by the drainagesystem. These stream channels expose the fault and exhibit aconspicuous shear zone with evidence of recent faulting(Fig. 4b). On the eastern block, the drainage system consistsof a small catchment area to the north and a single linear stream(E1) flowing from the east with a 60° N–80° N direction thatcuts into an abandoned alluvial terrace (Qt0). Because of thecatchment area, the northern bank of that stream has beeneroded andmodified, and only the southern bank is adequatelypreserved. Against the fault, stream E1 flows into a marshywater hole that may be partly man-made based on an existingdepression. On the western block, two strongly incisinggullies flow westward to the valley. The northern one (W1)continues west of the water pond (E1) along a 60° N directionand displays a subtle offset of 7� 0:5 m. The southern one(W2), while being very well expressed, has no counterparteast of the fault where a small catchment area has been formedby regressive erosion. Hence, the only possible source forW2is E1, making W2 a beheaded remnant left-laterally offset by114� 5 m. BecauseW1 andW2 cut into the upper surface ofthe Lisan formation, they necessarily are younger than itsultimate deposits; that is, they are younger than 25 ka B.P.(Abed and Yaghan, 2000). Furthermore, a trench exposure(see the Paleoseismology section, Fig. 4a, and Fig. 5) revealschannel deposits related to Qt0, which have been subse-quently radiocarbon dated at 19,700–16,800 B.C. (see sampleL-02 AkR fraction in Table 1). Considering that sample L-02originates from the middle of the stratigraphic section (seeFig. 5), the associated age is a minimum value for the empla-cement of the channel,which suggests aminimumageof 22kafor W1 and W2. Additionally, because the drainage source islimited to the surface of the late Lisan terrace, we may adoptthe approach developed by Ferry et al. (2007) in the same re-gion and assume the inception of that drainage was triggeredby an abrupt lake-level drop of Lake Lisan. The present eleva-tion of the terrace at Tell Saidiyeh is ∼255m below sea level,which, from the lake-level curve of Bartov et al.(2002) andinferences by Ferry et al. (2007), yield an age of 21–25 kaB.P. for the latest level drop below that elevation and indepen-dently confirmsour age inference. Taking into account the dat-ing of channels and terraces, we infer that the total left-lateraloffset of 114� 0:5 m has been accumulated during the last22–25 ka. The resulting long-term average slip rate is 4:9�0:3 mm=yr for that period, which is in good agreement with aprevious geological slip rate obtained from 20 offset streams(Ferry et al., 2007).
Instrumental Seismicity
The Dead Sea fault exhibits scarce instrumental seis-micity with mostly low to moderate earthquakes (M <6,
42 M. Ferry, M. Meghraoui, N. Abou Karaki, M. Al-Taj, and L. Khalil
Figure 3. Geomorphology of the Jordan Valley fault. (a) Central section of the JVF (see location on Fig. 1b), showing drainage (outlined)that is systematically left-laterally displaced at the passage of the fault. The active trace of the JVF is pointed out by the arrows. (b) Geo-morphology of the Tell Saidiyeh site from a high-resolution total station topographic survey (contour spacing 0.5 m). South of the archae-ological tell (located ∼100 m to the north, see inset in Fig. 8b), the morphology displays a recent terrace strath (Qt0) affected and left-laterallydisplaced by the fault. The southern edges (dashed lines) of streams serve as piercing points because they are less likely to be eroded than thenorthern ones in a left-lateral setting. Stream E1 flows westward along the southern edge of Qt0 and is displaced by 7� 0:5 m across thefault. Stream W2 is a beheaded remnant of E1 and displays 114� 5 m of offset. A minimum emplacement age of 22 ka for W2 yields anaverage slip rate of 4:9 mm=yr for that period (see text for details). Solid rectangles represent trenches T3 and T4 (see text for descriptions),which display faulting evidence for the last 17 ka. Blanked areas could not be surveyed due to the presence of agricultural and militaryfacilities. (c) Geomorphology of the Ghor Kabed site from a high-resolution total station topographic survey. The eastern fault strand shows alinear and continuous geometry with a gentle slope (the steep slope visible to the north is artificial), while the western strand displays a steeperslope and a left-step geometry. Two trenches were excavated at that site: T1 on the central strand north of the depression, and T2 on the easternstrand southeast of the depression (see logs in Fig. 5). Height curve spacing is 0.25 m. (d) The Ghor Kabed site displays a very subtlemorphology that could not have been fully deciphered without high-resolution topography. The color version of this figure is availableonly in the electronic edition.
Episodic Behavior of the Jordan Valley Section of the Dead Sea Fault 43
Figure 4. Field photographs of the Tell Saidiyeh and Ghor Kabed sites showing the shear zones (thick lines). (a) Oblique view oftrench T4 showing Lisan units affected by faulting and liquefaction in the foreground and sand and silt units in the background.(b) The main shear zone is outlined by a 1-cm-thick layer of crushed sand (thick lines) and displays numerous oriented pebbles. (c) Seismitesaffecting aragonite and detritus layers from Lisan units as observed at a nearby roadcut. (d) Main shear zone in trench T2 affectingLisan (massive clay) and Damya (clastic) units. Thin lines are stratigraphic contacts, thick lines are faults, and dashed lines mark thebottom of the trench. (e) Main shear zone in trench T1. (Legend as in part d.) Letters in parentheses in (d) and (e) refer to units in Figure 5.The color version of this figure is available only in the electronic edition.
44 M. Ferry, M. Meghraoui, N. Abou Karaki, M. Al-Taj, and L. Khalil
Figure5.
Trenchlogs
with
lithologicald
escriptio
ns.(a)
TrenchT1show
sadistributedpattern
ofverticalfaultsthatmay
beresolved
with
intheupperm
ostlayersbutcannotb
efollo
wed
throughmassive
clay
units
ofLisan
age.Radiocarbon
datessuggestthe
mostrecentevent
occurred
before
A.D.1490–1800.(b)
TrenchT2displays
amainfaultzonefilledwith
brecciathat
have
been
ruptured
afterw
ardanddocumentsthemostrecentevent,radiocarbon
datedafterA.D.5
60–6
60.C
ombined,
theseobservations
suggesttwosurface-rupturingeventsoccurred
atGhorK
abed
betweenA.D.560
andA.D.1800,which
may
berelatedto
theA.D.749
andA.D.1033events.(c)The
exposureof
T3ismainlycomposedof
Lisan
sediments.A
series
offine-
gained
collu
vialandalluvialunits
overlays
Lisan
claysandprovides
insightonrecentevents.(d)
TrenchT4isoriginallyaroad
cutthatw
asnoticeablyextended
andcleaned.Itisoriented
∼45°
tothefault,which
widensthedeform
ationzone.T
hisexposure
provides
thebulk
ofthepaleoseism
icdataset.Seetext
fordetails.(e)
Correlatio
nsof
stratig
raphicsections
oftrenches.T
hegeological
form
ations
ofLisan
andDam
yaarecommon
basement-botto
munits
fortrenches.E
rosion
processes(tild
elin
es)have
major
effectson
softsediments,and
TrenchT3show
sa
significanthiatusof
theDam
yaform
ation.The
correlationbetweenalluvialandlacustrine
depositsandtherelatedradiocarbondatin
g(see
also
Table1andFig.7)
illustratethedifferentrecent
depositio
nalenvironm
ents
attrench
sites.The
colorversionof
this
figure
isavailableonly
intheelectronic
edition.
(Contin
ued)
Episodic Behavior of the Jordan Valley Section of the Dead Sea Fault 45
Fig. 2a) mainly concentrated along the Lebanese Bend andthe Jordan Valley fault with three recorded moderateearthquakes (11 July 1927 Mw 6.2; 23 April 1979 Mb 5.2;and 2 February 2004 ML 5.2). However, the DSF is capableof producing large destructive events, as attested by the 22November 1995 Mw 7.3 Aqaba earthquake (Hofstetter,2003) and by large historical events (Fig. 2b; Ambraseysand Jackson, 1998; Sbeinati et al., 2005).
The 11 July 1927 event is the most destructive earth-quake to strike the region in the last century, with a knowntoll reaching 285 people killed and ∼1000 injured. Wide-spread destruction was documented by Willis (1928) in
Amman, Ramallah, Nablus, the Mount of Olives (“a mileeast of Jerusalem”), Reineh (Nazareth), As Salt, and Jericho(Fig. 6 for location). The event was recorded at more than100 seismological stations throughout the world, and itsepicenter was located within the northern basin of the DeadSea (Shapira et al., 1993). On the basis of reinterpretedhistorical documents, Avni et al. (2002) confirm these find-ings and describe a seiche wave in the Dead Sea that pleadsfor either a submarine landslide or a displacement of bathy-metry along a surface rupture. Submersible images of thebottom of the Dead Sea (Lazar and Ben-Avraham, 2002)show an apparently fresh and sharp scarp continuing the
T1
a
dc
e
f
ghijk
l
11200-12000 BC
AD 1490-1800
1320-1520 BC450-790 BC
AD 560-660
14700-16300 BC
modern
T2
a
d
c
e
f
g
b1b2
AD 1490 - 1640
T3
b
dc
ef
g
a1-3
T4
j
l
m
n
o
p
q
k1
i1
ab dce
f gh
12170-11720 BC11790-11310 BC
7500-4400 BC
10200-8800 BC
1610-1410 BC
5470-5110 BC5060-4770 BC
AD 1690-1920(AD 87-319)
AD 1660-1950
(17655-16475 BC)
11600-10910 BC11460-11270 BC
Dam
yaL
isan
Dam
ya
Dam
ya
Lis
an
Lis
an
~~~~~~
~~~~~~
~~~~~~
i2
k2
~~~~~~ ~~~~~~
1 m
~
~
~
(e)
Figure 5. Continued.
46 M. Ferry, M. Meghraoui, N. Abou Karaki, M. Al-Taj, and L. Khalil
Jordan Valley fault into the Dead Sea. It should be consideredthat the 1927Mw 6.2 earthquake may have produced surfacerupture locally with tenuous displacement, in the range of afew tens of centimeters (Wells and Coppersmith, 1994).
The 23 April 1979 event (epicenter 31.24° N, 35.46° E)was extensively instrumentally and macroseismically studiedby Arieh et al. (1982). It reached Io � IV–V MSK (thoughisoseismal lines are available for Israel only) in the JordanVal-ley and according to Arieh et al. (1982), its focal mechanismpoints to a 20° N-striking border fault of the northeasternDead
Sea basin. The 2 February 2004 event (epicenter 31.69° N,35.58° E) was felt strongly in Jordan, Israel, Palestine, andSyria and produced slight damage in Jordan and Israel, injur-ing a total of 20 persons (Jordan Seismological Observatory,2004). The combined analysis of aftershock distribution andfault plane solution suggests a normal fault branch perpendi-cular to the JVF, probably associated with the Dead Seapull-apart (Al-Tarazi et al., 2006). One may notice that theinstrumental seismicity does not reflect the level of active de-formation of the JVF and its potential for large earthquakes.
Figure 6. Archaeology of the Jordan Valley. White squares, main populated areas cited in historical documents; white dots, archae-ological sites visited and reappraised in this study; gray dots are archaeological sites not studied here (lack of evidence and/or availableliterature) but of potential interest for future studies; gray squares, paleoseismic sites; black squares, geomorphological sites studied by Ferryet al. (2007). The color version of this figure is available only in the electronic edition.
Episodic Behavior of the Jordan Valley Section of the Dead Sea Fault 47
Historical Seismicity
The historical seismicity relies on inscriptions and docu-ments from Greek, Hebrew, Roman, Byzantine, Arabic, andOttoman times (Guidoboni et al., 1994; Ambraseys, 2009). Awealth of testimonies, invoices, and reports are available forthe last twomillennia and document one of the most completehistorical catalogs to date. However, it should be noted that theregion was not evenly populated at any time in the past, withdensely populated areas along the Mediterranean coast andthe shores of the Sea of Galilee and the Dead Sea and barrenareas in the Negev desert and Wadi Araba. This situationinduces a systematic bias into intensity maps by (a) shiftingepicenters northward and (b) restricting the extent of felt tes-timonies. Furthermore, magnitudes derived from historicalstudies are usually very delicate to assess and carry a large(and undefined) uncertainty, especially for older events.Recent findings by Katz and Crouvi (2007) show that anthro-pogenic tali of archaeological origin have very poor geotech-nical characteristics and may amplify seismic shaking.Previous works suggest that MS >8 historical earthquakespossibly occurred in the Jordan Valley (Ambraseys and Jack-son, 1998). However, the length of fault segments andthickness of the seismogenic crust suggest that Mw 7.2–7.4is a reasonable maximum magnitude in this region.
Based on historical seismicity catalogs and recent litera-ture, we list here the main earthquake events that occurred inthe Jordan Valley (Fig. 2b):
• Event ZH (A.D. 1033, 10 Safar 425 A.H.): According to IbnAl-Jawzi (A.D. 1113–1200), on that morning the walls ofJerusalem crumbled down during construction, and thecities of Jerusalem, Ramallah, Jericho, Nablus, andTiberias were heavily damaged (Abou Karaki, 1987). Thisevent was felt throughout Judea and possibly as far away asEgypt and Syria, and it produced a sea wave along theMediterranean coast (Fig. 6 for locations; Poirier andTaher, 1980; Ambraseys et al., 1994).
• Event YH (A.D. 749): Theophanes (A.D. 760–818), a his-torian whose work constitutes one of the main sources forthat period, narrates “a powerful earthquake in Palestine,along the river Jordan and throughout Syria, and countlessthousands of people were killed, and churches and mon-asteries also collapsed, especially in the desert near theHoly City [Jerusalem]” (Ambraseys, 2009, p. 232). Thisevent has been intensively studied by numerous authors,due in great part to inconsistencies between calendars(Abou Karaki, 1987; Tsafrir and Foerster, 1992).
• Event XH (759 B.C.): This earthquake produced greatdestruction and many casualties in Judea, Samaria, andGalilee (Guidoboni et al., 1994). A thorough reappraisalof this event by Ambraseys (2005) indicates that few con-temporary accounts are available for this event, the earliestone being the Book of Amos. The first detailed descriptionis given by Zachariah around 520 B.C. (i.e., ∼240 yearslater) and suggests that a large landslide developed on
the Mount of Olives, southeast of Jerusalem, without aclear causative link.
Additionally, numerous other earthquakes have been feltin the Jordan Valley in historical times but may not be con-sidered as candidates for surface-rupturing events alongthe JVF:
• One may consider a candidate earthquake in A.D. 418 (notA.D. 419, as justified by Ambraseys, 2009). However,evidence is very weak because contemporaneous chroni-clers (Marcellinus Comes, Philostorgius; see also Ambra-seys, 2009) and archaeological investigations (Meyerset al., 1976) describe earthquake damage in the north ofGalilee region. There is no mention of major damage orvictims in Jerusalem, Jericho, and Palestinian villageslocated in the Dead Sea vicinity in A.D. 418.
• The A.D. 363 earthquake is better documented and consistsof a sequence of two shocks on 18 and 19 May (Ambra-seys, 2009). Although a large number of sites in Palestine,including Jerusalem, were damaged and a sea wave wasobserved in the Dead Sea, further south half of Petrawas razed to the ground, and localities near the Red Seawere badly damaged (Niemi and Mansoor, 2002).The wide region of damage and the ∼250-km-long WadiAraba fault from the Dead Sea to the Gulf of Aqabato the south may well be the site of two major shocksof A.D. 363.
• Historian Flavius Josephus (A.D. 37–100) vividly describesan event in 31 B.C. in the Jewish Wars: “For in the earlyspring, an earthquake shock killed an infinite number ofcattle and 30 thousand people; but the army was unharmed,because it was camped in the open” (Guidoboni et al., 1994,pp. 173–174). A critical and exhaustive reappraisal of thatevent by Ambraseys (2009) suggests that Flavius Josephus’account—the only coeval source—is greatly exaggerated,with a number of casualties larger than the actual populationof the region. The author also concludes that previouslyreported damage to archaeological structures is actuallyspurious and not associated with earthquake faulting. Insummary, a small to moderate earthquake probably oc-curred in 31 B.C. but did not produce significant damageto buildings or surface rupture. Asmentioned byAmbraseys(2009, p. 101), “The reappraisal of the available data revealsnothingmore than that the 31 B.C. earthquake in Judaea thatcaused damage and loss of life, which Josephus grossly ex-aggerates. There is no evidence that Jerusalem was affectedand the destruction or damage of other historical sites in Ju-daea is conjectural and cannot be tested on archeologicalground. The association of the earthquakewith a fault breakat Khirbet Qumran seems to me untenable and I can find nojustification for the addition of Diospolis to towns affectedand the dating of the event to A.D. 31 (Guidoboni et al.,1994; Guidoboni, 1989).”
• Previous studies consider an event in 64 B.C. that wouldhave damaged the Temple in Jerusalem and been feltthroughout the region and as far away as Antioch (south-
48 M. Ferry, M. Meghraoui, N. Abou Karaki, M. Al-Taj, and L. Khalil
east Turkey). However, the critical reinterpretation byKarcz (2004) strongly suggests an opposite situation, withan event source near Antioch (probably along the Hacipasafault or the Karasu fault) and related seismic shaking felt asfar as Jerusalem. This is supported by the fact that onlyminor damage was reported in the Jordan Valley. Indeed,this event is most likely the 65 B.C. Antioch earthquakethat claimed some 170,000 lives (Guidoboni et al.,1994; Sbeinati et al., 2005).
Paleoseismology
In order to establish a correlation between seismites andhistorical large events, Marco et al. (1996), Ken-Tor et al.(2001), and Migowski et al. (2004) have taken advantageof varvelike deposits in the Dead Sea around the southerntip of the JVF and the northern tip of the Wadi Araba fault.These authors claim an almost complete record of M >5:5earthquakes for the last 50 ka. However, due to erosionand depositional hiatuses, Ken-Tor et al. (2001) could notidentify the A.D. 1033 and A.D. 749 events in their varvesection. These events, as well as the 31 B.C. and 759 B.C.events, were identified by Migowski et al. (2004) in adifferent varve section. Furthermore, Migowski et al.(2004) identify three additional events in ∼1100 B.C.,∼2100 B.C., and ∼2700 B.C. (labeled events 36, 42, and43) for which historical data must be supplemented with ar-chaeology and paleoseismology.
Paleoseismic studies bring evidence for surface rupturesthat can be correlated with historical and prehistorical events.While historical and instrumental data do not point to aspecific seismic source, active faulting studies and paleoseis-mology may provide a direct observation of the causativefault. In order to perform successful paleoseismological inves-tigations, we carefully selected trench sites to ensure an opti-mal expression of faulting events, a continous and detailedsedimentary record, and material suitable for age determina-tions. In the following subsections of this paper, we describefour trench exposures (Fig. 5), for which deposit chronologiesare constrained by 28 radiocarbon samples (Table 1and Fig. 7).
Two trenches were dug across the fault scarps that limitthe northernmost pull-apart basin along the fault at GhorKabed (Fig. 3). Trenches are east–west-trending and dugacross the eastern fault section (trench T1) and across thewestern fault section (trench T2). The trenches expose thefault zone and related lacustrine (Lisan) and clastic (Damyaformation and Holocene) stratigraphic units. The stratigraphyin the two trenches (Fig. 5) is similar andmadeof (1) laminatedgray detritus (unit l in T1 and f in T2) visible at the trenchbottom, which can be correlated with the upper Lisan forma-tion; (2) a succession of 1.2-m-thick intercalated sandyand detritus layers (units k–g in T1) that corresponds to theDamya formation; (3) silt and sand units (units f–c in T1and f, d, and c in T2) that belong to the Holocene pull-apartdeposits; and (4) mixed units with detritus, silty-sand, and
sand (unit b in T1 and T2) visible mainly in the shear zonesoverlain by scattered caliche and organic soil (unit a in T1and T2).
Trench 1
In trench T1 (Fig. 5a), ruptures are distributed over thesection east of the main fault zone (Fig. 4e) and affect Lisanand Damya deposits. All upward fault terminations corre-spond to the base of the present-day plow unit (unit a)and do not show clear indications for a chronology. However,at the contact between Lisan/Damya and Holocene deposits,the faulted units correspond to a narrow fissure filled bypieces of unit b. Unit a, which covers the shear zone andcorresponds to an organic soil, has been dated at A.D. 1490–1800, postdating the most recent faulting event ZT1, whichpossibly corresponds to the A.D. 749 or the A.D. 1033earthquakes.
Trench 2
In trench T2, the surface rupture consists of several faultbranches in a 2.5-m-wide shear zone (Fig. 4d and Fig. 5b)that shows ∼1 m of total apparent vertical separation, withthe western block being the footwall. Here, abutting relation-ships permit the identification of four events:
• Event ZT2: The most recent event observed in trench T2 isassociated with surface ruptures that affect finely lami-nated unit b2, unit b1, and possibly c, d, and e with faultsplays terminating at the base of the top unit a. This event isnecessarily younger than unit b1, radiocarbon-datedA.D. 560–660, and may be associated with the historicalA.D. 749 earthquake and/or the A.D. 1033 earthquake.
• Event YT2: The event is attested by the formation of a1.5-m-wide flower structure filled with breccia (b1) andstratified silty clay (unit b2). In case unit b1 is a fissurefill, the event would have taken place shortly before thedeposition of unit b1 (i.e., shortly before A.D. 560–660).However, if unit b1 is composed of preexisting layersaffected by this event, it may then have occurred afterthe deposition of unit b1, which would naturally pointto the A.D. 749 earthquake. In that latter case, event ZT2
would correspond to the A.D. 1033 earthquake.• Event XT2: This event is documented by two fault splaysaffecting unit d up to the base of unit c, as well as by a∼40-cm-long vertical liquefaction dyke affecting unit dwith its source in the underlying sandy unit f. The scarcityof available datable material does not allow us to date thatevent accurately other than earlier than sixth century A.D.
• Event WT2: The oldest event that may be observed intrench T2 is attested by a Y-shaped rupture affecting unitsg, f, and the base of unit e at the eastern end of the trench.This event could be contemporaneous with the depositionof unit e (Damya formation).
The burial of unit b1 and related shear zone by unita in the two trenches (Fig. 5e) indicates a bracket of
Episodic Behavior of the Jordan Valley Section of the Dead Sea Fault 49
Table1
Radiocarbon
Datingof
Samples
Collected
inTrenchesT1andT2(G
horKabed
Site)andT3andT4(TellSaidiyeh
Site)*
Trench
Sample
Nam
eLaboratory
Code
Depth
(m)
Material
Fractio
n†Amount
ofCarbon
(AoC
,in
mg)
δ13C(‰
)Radiocarbon
Age
(B.P.)
Calibrated
Date2σRange
‡Observatio
ns
T1
Ka-T3-C04
KIA
24314
0.15
Sprig
AkR
5.4
�16:06
280�
30
1490
1800
T1
Ka-T3-C05
KIA
24315
1.15
Charcoal
HA
2.9
�25:38
11600�
50
�12000
�11200
T1
Ka-T3-S0
2KIA
24317
0.45
Carbonate
Carb
1.2
�7:92
3160�
30
�1520
�1320
T1
Ka-T3-S0
3KIA
24318
0.25
Carbonate
Carb
0.8
�8:52
2500�
20
�790
�450
T2
Ka-T2-1N
KIA
24305
0.5
Charcoal
AkR
5.3
�26:16
1440�
20
560
660
T2
Ka-T2-2N
KIA
24306
0.15
Peat
AkR
5.8
�23:63
Modern
--
T2
Ka-T2-6N
KIA
24309
1.5
Dustlayer
AkR
0.3
�24:85
14570�
250
�16300
�14700
Low
carbon
content;subject
tocontam
ination
T2
Ka-T2-9N
KIA
24311
0.2
Organic
mat
AkR
4.7
�25:67
Modern
--
T3
Tol-01
KIA
29719
0.3
Charcoalpieces
insand
AkR
3.4
�23:55
335�
20
1490
1640
T3
Tol-14
KIA
29720
0.2
Woodpieces
insandysoil
AkR
0.2
Modern
--
Mixture
ofprebom
bandpostbombcarbon
T4
L-02
KIA
29707
3.1
Carbonate
pieces
insandysoil
Carb
1.5
�4:35
15920�
70
�17655
�16475
Age
difference
between
CarbandAkR
fractio
nsnotstatistically
significant;
confirmsquality
ofthedate
T4
L-02
KIA
29707
3.1
Carbonate
pieces
insandysoil
AkR
0.2
�24:59
16950�
570=�530
�19700
�16800
T4
L-06
KIA
29715
2.1
Charcoaldust
insand
AkR
0.5
�29:81
11630�
120
�11790
�11310
T4
L-07
KIA
29708
2.2
Charcoaldust
insand
AkR
0.6
�28:40
12010�
100
�12150
�11720
Age
difference
between
AkR
andHA
fractio
nsnotstatistically
significant;
confirmsquality
ofthedate
T4
L-07
KIA
29708
2.2
Charcoaldust
insand
HA
2.4
�25:36
12125�
50
�12170
�11880
T4
L-14
KIA
29701
0.3
One
smallseed
ofcharcoal
piece
AkR
0.4
�21:41
118�
55
1660
1950
14C
ageplateau;
calib
ratio
nisundeterm
ined
T4
L-15
KIA
29702
0.3
Seed
(pum
pkin)
embedded
inthesoil
AkR
4.5
�25:87
Modern
--
Mixture
ofprebom
bandpostbombcarbon
T4
L-21
KIA
29717
1.9
Charcoaldust
insand
HA
2.9
�26:00
11455�
45
�11460
�11270
T4
L-22
KIA
29718
1.9
Charcoalpieces
and
dust
insand
AkR
0.8
�29:20
11040�
70
�11150
�10910
T4
L-22
KIA
29718
1.9
Charcoalpieces
and
dust
insand
HA
4.1
�24:54
11560�
50
�11600
�11320
T4
Tbc-04
KIA
30479
2.1
Sand
with
charcoal
pieces
AcidR
0.3
�27:31
9940�
180
�10200
�8800
Low
carbon
content;
subjectto
contam
ination
T4
Tbc-16
KIA
29709
0.2
Sand
with
plant
AkR
3.8
�23:34
50�
25
1690
1920
Age
difference
between
AkR
andHA
fractio
nsnotstatistically
significant;
confirmsquality
ofthedate
T4
Tbc-16
KIA
29709
0.2
Sand
with
plant
HA
5.0
�23:54
85�
20
1690
1919
T4
Tbc-16
KIA
29709
0.2
Smallsnail
Carb
2.5
�4:83
1825�
30
87319
Significantly
oldersnailshell,
probably
reworked
T4
Tbc-18
KIA
30480
1.7
Sand
with
charcoal
pieces
AkR
0.04
n.a.
7000�
700
�7500
�4400
Low
AoC
,norm
alized
tospecially
prepared
OxII-targets
T4
Tbc-23
KIA
29711
0.9
Bulksandysoil
AkR
0.6
�23:87
6015�
55
�5060
�4770
Low
carbon
content;subject
tocontam
ination
(contin
ued)
50 M. Ferry, M. Meghraoui, N. Abou Karaki, M. Al-Taj, and L. Khalil
Table1(Con
tinued)
Trench
Sample
Nam
eLaboratory
Code
Depth
(m)
Material
Fractio
n†Amount
ofCarbon
(AoC
,in
mg)
δ13C(‰
)Radiocarbon
Age
(B.P.)
Calibrated
Date2σRange
‡Observatio
ns
T4
Tbc-24
KIA
29712
0.7
Bulksandysoil
AkR
0.6
�26:82
6320�
60
�5470
�5110
Low
carbon
content;subject
tocontam
ination
T4
Tbc-26
KIA
29714
0.3
Bulksandysoil
AkR
1.7
�24:81
3210�
35
�1610
�1410
*Detailedmeasurementspresentedheregive
properinsightsregardingthequality
(amountof
carbon
shouldbe
largerthan
1mg)
ofcollected
samples,the
possibilityof
moderncontam
ination(depthandfractio
n),
andtheadequacy
ofcalib
ratio
n(δ
13C).
† AkR
,alkaliresidue;
HA,humic
acids;
Carb,
carbonate;
AcidR
,acid
residue..
‡The
2σcalib
ratio
nswereperformed
usingtheOxC
al3.10
software(Bronk
Ram
sey,
1995)andtheIN
TCAL04
calib
ratio
ncurvefrom
Reimer
etal.(2004).
Episodic Behavior of the Jordan Valley Section of the Dead Sea Fault 51
A.D. 560–1800 of the last two faulting movements inthe pull-apart area. Our interpretation is that the two post-sixth century faulting events may be correlated with theA.D. 749 and 5 December 1033 large earthquakes in theJordan Valley (Abou Karaki, 1987; Ambraseys and Jackson,1998).
Two further excavations were opened at Tell Saydiyehoutside of the archaeological perimeter (see Fig. 3 forlocation), across the main fault scarp and into an alluvialterrace.
Trench 3
Trench T3 was excavated perpendicular to the faultacross a 2-m-high scarp used for agricultural purposes(Fig. 3b). It exhibits a shallow network of subvertical faultsspread over 2 m and affecting Lisan deposits overlain with athin (<60 cm) succession of Holocene units. The bottom ofthe trench (Fig. 5c) is composed of laminated aragonite anddetritus (unit g) typical of Lisan deposits. Stratigraphy ismarked by the conspicuous aragonite layers and may be
15000CalBC 10000CalBC 5000CalBC CalBC/CalAD
Calibrated date
AD 1490-1800
BC 450-790
BC 1320-1520
Ph
ase
Tb
c-16
AD 1690-1920AD 1690-1919
AD 1660-1950
BC 1610-1410
BC 5060-4770
BC 7500-4400
BC 10200-8800
BC 11460-11270
Ph
ase
L21
-22
BC 11150-10910
BC 11600-11320
BC 11790-11310
Ph
ase
L-0
7
BC 12170-11880
BC 12150-11720
BC 5470-5110
AD 560-660
AD 1490-1640
BC 14700-16300
Ph
ase
L-0
2
BC 17655-16475BC 19700-16800
BC 11200-1200
AD 87-319
20000CalBCA
D74
9A
D10
33
ZT2YT2
ZT3
YT4XT4WT4Events VT4UT4
TT4
ST4AT4
BT4CT4
759
BC
1150
BC
2300
BC
2900
BC
depositional hiatus
pdf of calibrated date
pdf of event
hist./arch. event
Tren
ch 1
T2
T3
Tren
ch 4
Samples
Figure 7. Distribution of radiocarbon dates used in trenches 1–4 with inferred events, know historical earthquakes, and inferredarchaeoseismic events. Gray boxes indicate depositional hiatuses where no date could be determined. All dates given herein and in Table 1correspond to 2σ (95.4%) intervals on these probability density functions (pdf). Event pdfs are modeled for a Gaussian distribution on thebasis of inferred uncertainties defined in Table 3.
52 M. Ferry, M. Meghraoui, N. Abou Karaki, M. Al-Taj, and L. Khalil
resolved with difficulty when only detritus is present. Unit gis overlain with three groups of fine-grained deposits. A firstgroup is located on the central section of the trench and iscomposed of (1) a ∼20-cm-thick layer of siltstone (unit f),(2) a 5-to-15-cm-thick brown paleosol (unit e), and (3) an8-to-25-cm-thick layer of cemented silty sand (unit d). To thewest, a second unit truncates all other units and is composedof reworked material where remains of a wooden-soled boot,rust-encrusted hinges, and large lumps of charcoal werefound. It is unclear whether this anthropic unit is backfillfrom a small prior excavation or fissure fill. Indeed, the edgesof the unit do not display obvious signs of shearing, and noapparent vertical displacement is observed across that zone.The whole section is covered with a 20–30-cm-thick plowzone (unit b). At the toe of the scarp, all units are truncatedby a ∼1-m-wide man-made channel composed of two layersof clay (units a3 and a2) at the bottom and a 40-cm-thicklayer of clean gravels (unit a1) on top.
Because the scope of our study is limited to the mostrecent continuous record of events, we opted to focus onHolocene deposits and did not investigate Lisan units indetails. Thus, we could identify two events in trench T3.
• Event ZT3: This event affected unit e in two places, whichare east and west of a large modern root (see Fig. 5c). Theabsence of unit e west of the two fault splays suggests thatvertical displacement was larger than 15 cm on each splay.Event ZT3 has likely occurred shortly before the depositionof unit d dated A.D. 1490–1640 and probably correspondsto the A.D. 1033 earthquake.
• Event YT3: The oldest event recognized in trench T3 ismarked by the faulting of unit f, the oldest non-Lisan unitobserved here. It has likely occurred between the deposi-tion of units f and e. However, because event ZT3 cutthrough the whole thickness of unit e while event YT3
affects it partially, we assume that event YT3 occurredcloser to the deposition of unit e and event ZT3 closer tothe deposition of unit d.
It should be noted that the artificial fill unit at the wes-ternmost end of trench T3 may provide evidence for an extraevent. Indeed, considering the magnitude of the 1927 earth-quake, its location (Avni et al., 2002), as well as apparentlyrecent surface breaks at the bottom of the Dead Sea (Lazarand Ben-Avraham, 2002), a surface expression cannot beruled out for that event, with as much as 10–20 cm of co-seismic displacement. It follows that the artificial fill unitcould be an associated fissure fill.
Trench 4
Trench T4 (see Fig. 3 for location) is actually a cutrealized during leveling works to extend a nearby field, asmentioned by the field owner, and has a northwest–southeasttrend (i.e., oblique to the fault). It was widened and cleaned,and the first meter of material (perpendicularly to the expo-sure surface) was removed from the whole section to avoid
possible perturbations (e.g., ancient cliff collapse causingartificial deformation, contamination of potential radiocar-bon samples, dense vegetation on the top surface). TrenchT4 (Fig. 5d) displays a very well-expressed fault zone affect-ing late Pleistocene and Holocene units. West of the mainshear zone (FZ1), the deepest unit (q) is composed of finelylaminated aragonite and very fine gray clayey sand that maybe attributed to the Lisan. Unit q is strongly affected by soft-sediment deformation, liquefaction (unit r), and minor fault-ing (lower part of unit q). The overlying unit p is composedof massive clay with occasional pods of gravelly sand anddisplays deformation bands. Following the detailed descrip-tion by Abed and Yaghan (2000) of late Quaternary depositsin the region, that specific stratigraphic contact may be attrib-uted to the transition between Lisan (unit q) and Damya (unitp) dated by Abed and Yaghan at 16–15 ka B.P.. Unit p isoverlain with a series of alluvial units (n, m, and l) that dis-play upward reverse grading. Unit n is composed of graycoarse sand with occasional pebbles and grades into sandagainst the fault zone. Unit m is orange-yellow coarse sandwith occasional large pebbles. Unit l is a clast-supported peb-ble conglomerate that displays strong imbrication. The grouplies unconformably against unit p along an erosional surfaceand forms the northwestern edge of an alluvial channel. It isitself overlain with a thin red paleosoil (unit c) that caps themain shear zone, an irregular dark clay unit (b), and a matrix-supported sandy conglomerate (unit a) that truncates all unitsand forms the surface.
Southeast of the main shear zone, Pleistocene units arerepresented by a limited remnant of unit p that is observedat the southeastern-most end of the trench. Its top surfaceis erosional and overlainwith a yellowish sandy clay unit (unito) that does not appear in the northern block. Unit o is overlainwith a regular 10-cm-thick unit (h) composed of yellow siltwith carbonate nodules that sits immediately underneaththe topsoil unit. Units p, o, and h are affected by conjugatefaults that display more than 50 cm of apparent vertical dis-placement. Northwestward, unit o crops out at the base of thetrench and is affected by a series of minor faults. The top sur-face of unit o is deeply cut into by subsequent units. Unit nforms a wedge against unit o that we interpret as the easternedge of the channel described previously. At the same level,the central part of the trench displays a different picture.Indeed, the lowermost deposit is unit m—units p, o, and ndo not crop out there and must be deeply buried—and is over-lain with unit l, which pinches out against a major fault splay(FZ3) to the southeast. It is then overlain with coarse-to-finecemented sand units k2, k1, and j that exist only there. Thecommon erosional top surface of units n and j is overlainby a group of sandy channel units (i1 and i2) that cut into unitso and n and are overlain by unit h. Those three units extendnorthwestward against themain shear zone (FZ1).While unitsn, i2, and i1 cannot be clearly followed inside of the shearzone, a 50-cm-long section of unit h can be observed withinit and is vertically offset by ∼40 cm. Unit h is overlain with agroup of subhorizontal fine-grained units (g–d), composed of
Episodic Behavior of the Jordan Valley Section of the Dead Sea Fault 53
varying amounts of yellow to brown cemented silt and clay.Capping the fault, unit c may be followed eastward for∼30 cm. It is then confused with the topsoil.
The different units are affected by a complex network offaults and fissures. The main fault zone (FZ1 in Fig. 5d) is30–70 cm wide and displays densely packed gravels (prob-ably incorporated from unit l) with the long axis of pebblesoriented subvertically along the main shearing direction(Fig. 4b). Where it affects sandy units, the main fault zone(FZ1) is outlined by a ∼1-cm-thick band of white pulverizedsand. Three supplementary fault splays can be traced fromthe bottom up to the shallowest units and are named FZ2,FZ3, and FZ4. They are ∼1 cm wide and filled with abrown-red silty clayey material that may originate fromthe uppermost soil units. Besides, a wealth of minor splaysand numerous fissures affect the central part of the section,between FZ1 and FZ3. It should be noted that the width ofthe fault zone is only apparent due to the obliquity of thetrench with respect to the fault’s direction. Though not fol-lowing standards of paleoseismic trenching, this situationdoes provide a better exposure and extended wall surfaceto collect more observations and samples for dating.
Starting with ZT4 as the most recent event, we identifieda set of eight to nine surface-rupturing events affectingalluvial Holocene and Damya units, as well as three olderevents affecting Lisan deposits (see circled letters in Fig. 5d).The lack of visible stratification in the massive clays of unit pprevents us from identifying deformation features and recon-structing the corresponding part of the faulting history.
• Event ZT4: This most recent event is illustrated by threemajor splays (FZ1, FZ3, and FZ4) that affect the wholestratigraphic section up to 20–30 cm below the present-day surface. Vertical displacement can only be resolvedon FZ3, where it reaches ∼5 cm. That rupture does notaffect the shallowest units b and c. It has likely occurredafter the deposition of unit d and before the deposition ofunit c, thus yielding a time window between A.D. 87 andA.D. 1920 (Table 1) and pleading for a historical event.Because the A.D. 87 lower bracket is based on a snail shellthat is significantly older than the surrounding soil, weconsider that the event occurred significantly closer to theupper bracket; that is, more likely after ∼A.D. 500.However, from the available radiocarbon datings alone,it is not possible to decide if this exposed fault has experi-enced rupture in A.D. 749 or A.D. 1033 or both. Alterna-tively, one may argue that unit c (dated A.D. 1660–1950)exhibits noticeable warping across the main fault zone withan apparent vertical deformation of ∼25 cm and that unit bthickens at the toe of the related scarplet into what may be acolluvial wedge. Age and dimensions of those featurescorrespond to a recent Mw ∼ 6 earthquake, such as the1927 Palestine earthquake. This interpretation is supportedby the occurrence of a modern fissure fill unit in T3.
• Event YT4: This event is interpreted from small (a fewcentimeters) displacements affecting units along FZ2.
All units in the central section from m to e display minoroffsets. Unit d caps the rupture and forms the eventhorizon. Event YT4 occurred between the deposition ofunits e and d and may be dated by samples Tbc-23 andTbc-26 (Table 1). This yields a wide window of occurrencebetween 5060 B.C. and 1410 B.C.
• EventXT4: This event is marked by a fan-shaped network ofsplays located immediately southeast of FZ2. Three splaysaffect units h, g, and f and produce ∼5 cm of cumulatedvertical throw. They are consistently capped by a thin, siltyclay unit. Event XT4 can be dated by bulk soil samplesTbc-23 and Tbc-24. The stratigraphically lower sample23 is dated 5060 B.C.–4770 B.C. and appears to be slightlyyounger than sample 24 (dated 5470 B.C.–5110 B.C.), thussuggesting an age inversion. However, samples 23 and 24only produced 0.6 mg of carbon and may therefore besubject to contamination. Because samples are bulk soil,contamination is probably related to exposure, manipula-tion, and storage conditions and should then be associatedwith a rejuvenation process. This would imply that theactual age of samples may be slightly older than the mea-sured ones. In summary, this suggests that sample 23 shouldbe somehow older and points to an occurrence shortlybefore the deposition of the thin, silty clay unit (i.e., between5470 B.C. and 5000 B.C.). EventXT4 may also be documen-ted by a fault splay located immediately west of FZ3, whichaffects the limit between units g and f. However, the faulttermination could not be pinpointed.
• Event WT4: This event is documented by a fault splaylocated ∼1 m northwest of FZ3. This splay affects all unitsup to unit h and is capped by unit g. Its occurrence timemay be bracketed by samples Tbc-23/Tbc-24 and Tbc-18and yields a window between 7500 B.C. and 5080 B.C.Considering the stratigraphic position of event WT4 withrespect to samples, we propose the actual date is closerto the age of sample Tbc-18 and is therefore probably com-prised between 7500 B.C. and 5500 B.C.
• Event VT4: This event is located on FZ3, a fault splay thatbroke during eventZT4. However, the bottom limit of unit i1displays about twice as much displacement as the bottom ofunit h, thus suggesting that an event took place between thedeposition of units i1 and h. This yields a probable time ofoccurrence between 11,600 B.C. and 4400 B.C. Consideringthe possible rejuvenation of sample Tbc-18 (total amountsof carbon [AoC] of 0.04mg; see Table 1) and the intermedi-ate stratigraphic position of event VT4, we estimate theoccurrence date between 10,900 B.C. and 7500 B.C.
• Event UT4: This event is located along the same fault zoneas event XT4 but ∼0:8 m deeper. In that part of the section,the faulting pattern is somehow more complex and moremature. Because the top limit of unit k2 is strongly shearedand cumulatively offset by ∼60 cm, we infer that an olderevent took place there after the deposition of unit k1. Faultterminations are unclear in that particular part but affectsome features that we correlate with the bottom limit ofunit i2. Upward, the splays cannot be resolved, and it is
54 M. Ferry, M. Meghraoui, N. Abou Karaki, M. Al-Taj, and L. Khalil
unclear whether they affect unit i1 or not. Several splaysseem to stop within unit i2 (see splays left of event symbolU in Fig. 5d), suggesting event UT4 took place after thedeposition of unit i2. This yields a time window between11,600 B.C. and 10,200 B.C. Considering the stratigraphiclocation, we propose an event date between 11,500 B.C.and 10,500 B.C.
• Event TT4: This event is documented by a splay thatoriginates from the main fault zone FZ1 and displays anapparent dip of ∼45°. This splay strongly affects unit l(gravels) and displaces the bottom of unit k1 vertically by20 cm and its top surface by only a few cm, and it maybe followed upward 50 cm into the base of unit j. The as-sociated event may hence be postdated by sample L-02.However, this sample has a relatively low AoC (mg), iscomposed of carbonate (susceptible to be reworked), andappears to be older than the stratigraphically lower andwell-dated samples L-07 and L-06. Thus, we choose not to relyon sample L-02. Consequently, we propose that eventTT4 isbracketed by samples L-21/L-22 and L-06, which yields anoccurrence date between 12,060 B.C. and 10,910 B.C.
• Event ST4: A conspicuous group of minor fault splaysaffect a clearly defined subhorizontal layer within unit oin the southeast section of the trench, thus indicating thatevent ST4 occurred after the deposition of unit o. However,due to subsequent erosion of that unit and the emplacementof younger channel deposits, all fault splays have been cutand stop at the erosion surface. Considering that the olderchannel composed of units n, m, and l has originally cutinto units o and p (and totally removed unit o from thewestern-most section) we may infer that event ST4 has ac-tually occurred prior to the deposition of unit n. This strongerosion process has erased part of the stratigraphic recordand limits the availability of dating samples. Consequently,event ST4 may only be defined as having occurred withinthe same time windows as event TT4; that is, between10,910 B.C. and 12,060 B.C. It may be assumed that themain channel fill is noticeably thicker than the exposedsection, which would put event ST4 stratigraphically closeto sample L-06. This would suggest that event ST4 oc-curred closer to the lower end of the bracket; that is,presumably between 12,060 B.C. and 11,500 B.C.
• Event CT4: This event is the only clear liquefaction eventthat was identified at that site. It is marked by a 0.5-by-1-m-large pocket of homogeneous, fine-grained, red, well-sorted carbonate sand surrounded by distorted detritusand aragonite layers. Furthermore, the pocket truncates anolder fault that may be attributed to an older event (seeEvent BT4).
• Event BT4: This event is pointed out by a single fault splaythat cuts through Lisan units and was later truncated byliquefaction from event CT4.
• Event AT4: The oldest event visible in exposure T4 affectsthe lowermost Lisan laminae as a fan-shaped splaynetwork. All splays, as well as a nearby seismite feature,are consistently truncated by a subsequent deposit.
Events CT4, BT4, and AT4 all occurred during the deposi-tion of Lisan units or shortly after, while they were still watersaturated. This dates all three events back to the end of theLisan (shortly before 20 ka B.P.; Bartov et al., 2002).
To achieve the best possible characterization of events,we took advantage of outcropping site-wide and region-widesedimentary formations (Fig. 5) and enhanced our analysisusing stratigraphic correlations across trenches at a given siteand across sites. Hence, our four paleoseismic trenches,opened at two sites along the central and southern sectionsof the JVF, yield a total of 12 surface-rupturing events: 2 maybe correlated to historical earthquakes (A.D. 1033 andA.D. 749), and the remaining 10 are prehistoric. The oldestevents identified (A, B, and C) are synchronous with thelatest Lisan units deposited between 17 ka B.P. and 20 kaB.P. It should be noted that, due to sedimentary hiatuses,no event could be identified between the last historical earth-quakes and YT4 (5060 B.C.–1410 B.C.), thus leading to asignificant gap in the record. Considering the proximitybetween the active trace of the JVF and archaeological sites,the adequate time period, and the rich record, we propose torely on archaeoseismology to compensate for the incompletepaleoseismic data.
Archaeoseismology
Rift valleys throughout the world are common passage-ways for human populations and are thus generally rich withthe archaeological heritage of former civilizations. The evo-lution and migration of human groups with respect to activefaults has hence long been established (King et al., 1994).Here, we first provide detailed evidence for earthquake-in-duced destruction at Tell Saidiyeh, an archaeological site thatsits a few tens of meters from the active trace of the JVF(Fig. 3) and which has been studied by means of paleoseis-mic excavations (trenches T3 and T4 in section 5). Addition-ally, we present a critical reappraisal of published data about20 archaeological sites scattered over the Jordan Valley andneighboring regions (Fig. 3 and Fig. 6), which show varyingdegrees of evidence for earthquake-related damage andpossibly surface rupture. The archaeological evidence forpaleoearthquakes takes the form of destruction to buildingswith frequent fires and consequent signs of site abandon-ment. Thus, a widespread burnt layer with a noticeablecontent of rubble, pottery shards, and ashes may be a goodcandidate for earthquake evidence. However, indication foran earthquake is generally reduced to signs of destructionthat can be instead related to regional wars, local raids, oreven accidents (e.g., accidental fire caused by an unattendedoil lamp). In a recent review, Ambraseys (2006) underlinesthe different caveats pertaining to the use of archaeologicalevidence to identify past earthquakes and particularly to theinterpretation of toppled structures. We follow Ambraseys’sguidelines in making our interpretations of existing archae-ological data from Tubb (1988 and 1998), Savage et al.(2001, 2002, and 2003), and Franken (1989) and retain the
Episodic Behavior of the Jordan Valley Section of the Dead Sea Fault 55
most robust evidence to identify destruction events that maybe attributed to past earthquakes.
For example, Tell Saidiyeh (Fig. 8) is located almostexactly at the center of the Jordan Valley (Fig. 6), whichsuggests the site would probably experience intense ground-shaking in the case of a major earthquake on the JVF. Tellstructures are common throughout the Middle East and arecomposed of layers of successive settlement (spreading overa few centuries to several millennia), which may pile up toreach 30–50 m elevation above the base level. Tell Saidiyehhas been studied for the last 60 years and been the object ofsystematic excavations since 1964 (Tubb, 1988). The site hasproduced a wealth of artifacts that document rather contin-uous occupation from the Chalcolithic (fourth millen-ium B.C.) up to the Roman period (Tubb, 1998). The tellitself is composed of two distinct mounds (Fig. 8), withnoticeably different histories: an upper main tell whereelaborate structures were discovered (e.g., an olive-pressingcomplex and a water staircase; see Fig. 8d) and a lower tellthat was mostly used as a burial ground during the Omayyadperiod. A compilation of indications for strong perturbationsto Tell Saidiyeh (Table 2) point to two specific stratafor which destruction is significant: stratum L2, dated∼2900 B.C., and stratum XII, dated 1150–1120 B.C. For bothstrata, the author mentions widespread damage with intenseburning, collapsed walls, and broken potteries (Fig. 8c). Anadditional event may be linked to stratum VI, dated to themiddle of the eighth century B.C., and may correspond tothe 759 B.C. Jericho earthquake (Nur and Cline, 2000).Indeed, Tubb (1998, p. 126) mentions that “houses of Stra-tum VI were knocked down and leveled in preparation foranother major building programme.” The reason for sucha drastic solution may be the prior intense destruction ofthe tell by an earthquake with no possibility to re-use da-maged buildings. Other strata show evidence for destructionor abandonment but could not be related to seismic shaking.
In parallel, we compile existing data for 11 archaeologi-cal sites in the vicinity of the JVF showing a potential for pastearthquake damage (Fig. 6 and Table 2). However, we con-sider earthquake-induced damage in archaeological sites onlyif it is attested by a minimum of two sufficiently distant sitesand sites with evidence for surface rupture. The analysis ofdamage to archaeological sites allows us to identify fourevents likely associated with large earthquakes along the JVF:
• Event ZA: This event is attested at Tell Deir’ Alla (Franken,1989) and Tell Saidiyeh (Tubb, 1988) for the middle of theeighth century B.C. Available descriptions lack details, andit is probable that the corresponding destruction has beenindirectly associated with the well-known 759 B.C. Zechar-iah’s earthquake (Nur and Ron, 1996) without further agedetermination. At Tell Saidiyeh, damage is not directly as-sociated with an earthquake, but it is rather the subsequentmassive leveling of the site that suggests a catastrophicevent. Proposed date: 759 B.C..
• Event YA: This event is attested at the neighboring sites ofTell Saidiyeh and Tell Deir’ Alla, where the occurrence ofan earthquake in the early twelfth century B.C. leaves nodoubt for Franken (1989) and J. N. Tubb (personal comm..2005). At Tell Al’ Umayri, ∼30 km east of the JVF,contemporary destruction associated with a burn layer richin broken vessels is documented by Savage et al. (2001). Itshould be noted that this event may be part of thewell-documented twelfth century B.C. “earthquake storm”studied by many authors and summarized by Nur and Cline(2000). Proposed date: ∼1150 B.C..
• Event XA: This event is documented at Tell Abu en-Ni’ajby Savage et al. (2003) as a series of ash layers offset by afault splay. According to our mapping of the JVF (Fig. 6;Al-Taj, 2000; Ferry et al., 2007), Tell Abu en-Ni’aj appearsto be west and off the main active trace by ∼1 km. Savageet al. (2003) do not provide details about the amount orintensity of deformation along that fault, and it is presentlynot possible to decide whether the observed offsets arerelated to coseismic fault slip. At Khirbet Iskander, located∼30 km southeast of the JVF (Fig. 6), Savage et al. (2003)describe a destruction layer in great detail (Table 2), wherefire was attested by burnt stones, ash layers, and charredgrain. Well-preserved wooden beams and human remainssuggest a sudden possibly earthquake-related catastrophe.Proposed date: ∼2300 B.C.
• EventWA: At Tell Saidiyeh, Tubb (1988) documents densedestruction debris associated with fragmentary and dis-turbed architecture and indicates these are consistent withground-shaking-related damage (J. N. Tubb, personalcomm., 2005). At the nearby Tell el-Fukhar site, Savageet al. (2003) mention evidence of intense destruction towalls and infer a possible earthquake-related origin. Pro-posed date: ∼2900 B.C.
Overall, our compilation of archaeological studies ontells in the vicinity of the JVF strongly suggests the occur-rence of four events: in 759 B.C., ∼1150 B.C., ∼2300 B.C.,and ∼2900 B.C.. Here, the archaeological record intersectsthe historical record, as attested by the 759 B.C. earthquakedescribed in written documents. Furthermore, the archaeo-seismic event XA can be correlated with the paleoseismicevent YT4 and compensates for the sedimentary hiatusobserved in paleoseismic trenches (period ∼1500 B.C. to∼4500 B.C.). It should be noted that indications for surfacebreaks affecting tells may not necessarily be considered asfaulting evidence. Indeed, tells are artificial mounds madeof heterogeneous material that includes rubble, dirt, andarchitectural remains (Fig. 8a). As such, they are very sen-sitive to gravitational collapse, especially under seismicshaking. Hence, for sites that are not directly located acrossthe main active trace of the JVF, we consider surface defor-mation as secondary evidence in the sense of McCalpin(1998). Nonetheless, such indications may provide a relevantchronological constraint for a seismic event. Thus, themention by Franken (1989, p. 203) of “a victim �…� found
56 M. Ferry, M. Meghraoui, N. Abou Karaki, M. Al-Taj, and L. Khalil
Figure 8. Tell Saydiyeh archaeological site. (a,b) General view of the site and satellite imagery (inset) showing its relationship to thefault. The archaeological site is composed with a lower and an upper tell that were occupied at different periods. The original site was alimited mound that looked over the region and has grown with the successive addition of settlement layers, each related to a specific period.The obvious proximity of the JVF is marked by the active fault scarp. In (b), the satellite imagery shows the relationship between the activefault trace (thick line), the archaeological site (LT, lower tell; UT, upper tell), and geomorphology (white outline represents the extent of themicrotopographic survey in Fig. 3b). (c) Open pit at the top of the tell showing conspicuous ∼5-cm-thick black burnt layers. Those layerscontain broken pottery, charred wood, and ashes and are remnants of a widespread intense fire. (d) The twelfth century B.C. olive processingarea (Palace) that displays signs of destruction (from Tubb, 1998). (e) Blocked doorway and broken vessel interpreted as a direct result ofearthquake shaking (from Tubb, 1988). The color version of this figure is available only in the electronic edition.
Episodic Behavior of the Jordan Valley Section of the Dead Sea Fault 57
Table2
Com
pilatio
nof
ArchaeologicalEvidenceforStrong
Perturbatio
nsat
ArchaeologicalSitesin
theVicinity
oftheJordan
ValleyFault
Docum
entedDate
Archaeological
Period
Dateof
Event
(Inferred)
Site
Descriptio
nfrom
ArchaeologicalSo
urce
Proposed
Cause
ofDestructio
nProbability
ofan
Earthquake
Indicatio
nof
Surface
Rupture
?Mid-fourteenth
centuryA.D.
∼A:D:1350
TellHisban
“There
isevidence
ofearthquake
collapseandfire
for
themid
14th-century
destructionthat
preservedthe
contents
ofthestore.”(p.446)*
Earthquake
High
?Ayyubid-M
amluk
A.D.1170–1520
TellYa’am
un“T
hemosaicfloorof
theeast
room
isextensively
dented
bycollapsed
wallstones,which
suggeststhat
useendedwith
destructioncaused
byan
earthquake.”(p.457)
†
Earthquake
ZH(749
A.D.)
A.D.749
Khirbet
Yajuz
“Inarea
E,anearthquake
thatoccurred
inA.D.7
48is
illustrated
bythecollapsed
vaultedarches
andthe
irregularitiesof
thepavedfloor,which
date
tothe
Umayyadperiod.”(p.448)
‡
Earthquake
High
?Mid-seventh
centuryA.D.
∼A:D:650
TellHisban
“After
anearthquake
inthemid
seventhcenturyA.D.,
which
was
responsibleforthecollapseof
thestone
barrel
vaults,the
structurewas
reoccupied
andused
into
theAbbasid
period.”(p.446)*
Earthquake
High
?LateByzantin
eA.D.490–640
Jerash
“The
pottery
andglassunderthistumbled
wallsectio
nshow
edthatthecollapsemusth
aveoccurred
during
theLateByzantin
eperiod,p
robablytheresultof
anearthquake
that
was
responsibleforthedestruction
ofothercity
build
ings
inthesixthcentury.”(p.458)†
Earthquake
ZA(759
B.C.)
Seventh–sixth
centuryB.C.
500–700B.C.
TellSaydiyeh
“Aseries
ofbuild
ingphases
(IIIB–IIIG)was
defined
belowPritchard’sStratum
III(now
term
edIIIA
),the
lowestof
which
show
sarchitecturevery
similarto
Stratum
V,with
similarevidence
forburning.”
(p.130)
§
Fire
Low
∼720B:C:
TellSaydiyeh
“These
stalls
werefrequently
foundto
containequid
bones,anditseem
slik
elythat
theserepresentthe
remains
ofunfortunateanim
alswhich
hadbeen
abandonedto
thefire
which
broughtan
endto
Stratum
Varound
720BC.The
destructionof
Stratum
Vmight
have
been
accidental,butitmight
also
beattributed
totheAssyrians,who
were
campaigning
inthis
region
atthetim
e.”(p.127)
§
Accidentalfire?
War?
Average
Eighth
centuryB.C.
700–800B.C.
Deir’A
lla“Phase
Mwas
destroyedby
earthquake
andfire.”
(p.204)
∥Earthquake,
fire
Mid-eighth
centuryB.C.
∼750B:C:
TellSaydiyeh
“Tow
ards
themiddleof
theeigthcenturythehouses
ofStratum
VIwereknockeddownandleveledin
preparationforanothermajor
build
ingprogramme.”
(p.126)
§
Anthropic
postearthquake?
Average
(contin
ued)
58 M. Ferry, M. Meghraoui, N. Abou Karaki, M. Al-Taj, and L. Khalil
Table2(Con
tinued)
Docum
entedDate
Archaeological
Period
Dateof
Event
(Inferred)
Site
Descriptio
nfrom
ArchaeologicalSo
urce
Proposed
Cause
ofDestructio
nProbability
ofan
Earthquake
Indicatio
nof
Surface
Rupture
YA(∼1
150B:C:)
1150–1120B.C.
TellSaydiyeh
“The
cityof
StratumXIIwas
obviouslydestroyedinan
intenseconflagration,
thedenseassociated
debris
sealingavaluable
corpus
offinds,exam
inationof
which
hasestablishedadateforthiseventataround
1150-1120BC,coinciding
with
thewith
draw
alof
theEgyptianem
pire...
Sometim
ein
thelastquarter
ofthe[twelfth]
centurythecity
ofStratum
XIIwas
destroyedby
fire,and
atthesametim
ethecemetery
ontheLow
erTellfelloutof
use.
There
isno
indicatio
nas
tothesource
ofthedestruction:
certainlytherewereneith
erbodies
norsignsof
conflictam
idst
theruined
build
ings
oftheUpper
Tell,
anditcouldwellbethatfirewas
theresultof
anaccident.”(p.86)§
Fire
High
Early
twelfth
centuryB.C.
1150–1100B.C.
Deir’A
lla“T
here
islittle
doubtthat
theentirecomplex
was
destroyedearlyin
the12th
c.BC
andthat
anearthquake
caused
thedestruction.”(p.203)
∥
Earthquake
LatefirstIron
Age
1200–1100B.C.
Deir’A
lla“T
hisperiod
endedagainwith
anearthquake
anda
victim
was
foundcompletelysquashed
inacrackin
theearth.”(p.203)
∥
Earthquake
Yes
Early
Iron
Age
1200–1100B.C.
Tellal-’Umayri
“Inan
adjacentbuild
ingtothesouth,destructiondebris
coveredalayerof
burned
andbroken
ceramic
vessels.”(p.440)
‡
?
?MiddleBronze
Age
∼1600B:C:
Tellal-’Umayri
“Anearthquake
distortedmanyof
thenorth–south
wallsfrom
thisperiod
atthesite,including
someof
thosein
thepalace.”(p.463)
†
Earthquake
XA(∼2
300B:C:)
Early
BronzeIV
∼2300B:C:
TellAbu
en-N
i’aj
“[...]bulld
ozingon
thewestern
side
ofTellAbu
en-
Ni'ajcleared
a26-m
-longstratig
raphiccross-section,
revealingan
earthquake
slip
fault.Early
BronzeIV
deposits
capped
aseries
ofoffset
ashlayers,
indicatin
gthat
theearthquake
occurred
during
the
occupatio
nof
TellAbu
en-N
i'aj.”
(p.439)
‡
Earthquake
High
Yes
Early
BronzeIII
2650–2350B.C.
Khirbet
Iskander
“[.....]
sectionclearlyrevealed
thetip
lines
ofburnt
stones,seriesof
ashlayers,m
udbricks,abd
detritu
s,clarifiedthattheseremains
represento
nedestruction
phase.
[...]
thedestructiondebris
[......]included
restorable
pithoi
andwavy-handledvessels,well-
preservedwoodenbeam
s,andquantitiesof
charred
grain,
lentils,andpeas.The
charredbonesof
analmostcom
pletehuman
armlayon
theplasterfloor.”
(p.541)
#
Earthquake?,Fire
(contin
ued)
Episodic Behavior of the Jordan Valley Section of the Dead Sea Fault 59
Table2(Con
tinued)
Docum
entedDate
Archaeological
Period
Dateof
Event
(Inferred)
Site
Descriptio
nfrom
ArchaeologicalSo
urce
Proposed
Cause
ofDestructio
nProbability
ofan
Earthquake
Indicatio
nof
Surface
Rupture
WA(∼2
900B:C:)
Lateearly
BronzeI
∼2900B:C:
TellSaydiyeh
“Stratum
L2was
foundto
beassociated
with
dense
destructiondebris
(ashes,burntmud-brick
rubble
andcharredtim
ber),but
both
StrataL2andL3were
apparently
built
onthesameplan.In
thecentrally
locatedarea,excavatio
nsrevealed
extrem
ely
fragmentary
anddisturbedEarly
BronzeAge
architecture,
theremains
having
been
allbut
obliterated
bythediggingof
graves
inthethirteenth
totwelfthcenturyBC.”(p.42)§
Unknown
High
Early
BronzeIB
3300–2850B.C.
Tellel-Fukhar
“The
terracewallswereso
damaged,p
resumably
from
earthquakesof
which
wefoundevidence,that
they
weredifficulttodistinguishfrom
thehuge
stonefalls
around
them
.”(p.462)
†
*Savageet
al.(2002).
† Savageet
al.(2003).
‡ Savageet
al.(2001).
§ Tubb(1998).
∥ Franken
(1989).
# Savageet
al.(2005).
60 M. Ferry, M. Meghraoui, N. Abou Karaki, M. Al-Taj, and L. Khalil
completely squashed in a crack in the earth” at Tell Deir’Alla(∼3 km east of the JVF) and Savage et al.’s (2003) previouslymentioned “earthquake slip fault” at Tell Abu en-Ni’aj arenot considered as strong evidence for surface rupture butmay document off-fault effects of a large earthquake.
Summary of Paleoseismic andArchaeoseismic Events
We determined 3 seismic events from historical studies(ZH–XH), 4 damaging events from archaeological studies(ZA–WA), and 12 faulting events from paleoseismic investi-gations (ZT2 and YT2, ZT3, ZT4–ST4, and CT4–AT4). The con-sistency in the correlation between historical, damaging, andfaulting events constrains the occurrence of the earthquakeevents. Table 3 presents the resulting catalog of 15 large earth-quakes (Z–O and C–A) produced by the JVF betweenA.D. 1033 and 12,060 B.C. and between 15,000 B.C. and18,000B.C. (Fig. 9). It should be noted that an additional eventwas identified in trench T2 (WT2) for which we could not es-timate an age. The fusion of the different datasets reinforcesthe identification of past earthquakes. Indeed, of the 13 paleo-seismic events, two are also present in the historical record andone (possibly two) in the archaeological record. One event ispresent in both historical and archaeological datasets.
Constraints on the Holocene Behaviorof the Jordan Valley Fault
With major barriers (the Dead Sea and the Hula basin) tothe rupture propagation toward nearby fault segments and
very weak structural relays (stepovers) within the JVF, itis likely that large earthquake ruptures are characteristic inlength and associated withMw 7.2–7.4 and a ∼3:3 m averagecoseismic displacement. An estimate of the total seismicmoment release for the 12 faulting events (Z–O in Fig. 9)during the last 14 ka yields an ∼3:3 mm=yr slip rate, whichprobably indicates a lack in the paleoseismic record (possiblydue to the sedimentary hiatus). By contrast, using the totalseismic moment for the admittedly most complete part of thecatalog (events Z–U) over a period of ∼3:9 ka, we obtain an∼4:2 mm=yr slip rate (Fig. 10), which is comparable to theslip rate obtained from offset streams and paleoclimatic fluc-tuations (Ferry et al., 2007). The slight difference is easilyexplained by distributed deformation around the main trace(possibly ∼0:5 m per event).
Taken as a whole, our integrated catalog of past seismicevents yields a mean recurrence interval of 1165 yr (standarddeviation 243 yr). A close examination reveals highly vari-able recurrence intervals with values ranging from 284 yr to2700 yr (Table 3). To some extent, we agree that very highvalues may reflect the incompleteness of our catalog,especially for its purely paleoseismic part (events T–O).However, long intervals are also present in the assumedlycomplete historical and archaeological parts, between eventsY and X (1508 yr) and events W and V (1150 yr), whichsuggests very long quiescence periods are real. In parallel,short interval values are clearly observed through the wholecatalog: between events Y and Z (284 yr), between events Wand X (391 yr), and across events O, P, and Q (190–520 yr).We propose that these short intervals reflect clustering and agenerally episodic behavior of the JVF over the last 14 ka.
Figure 9. Events probability density functions for the last ∼20 ka along the JVF from historical, archaeological, and paleoseismic data.The average recurrence interval is 1165 yr (σ � 243 yr) for the whole period but varies from 787 yr (σ � 1212 yr) for historical andarchaeological data to 1480 yr (σ � 705 yr) for paleoseismic data. See text for a detailed analysis. The color version of this figure is availableonly in the electronic edition.
Episodic Behavior of the Jordan Valley Section of the Dead Sea Fault 61
This independently confirms the observation by Ferry et al.(2007) of slip rate variations along the JVF from a long-termvalue of 4:9 mm=yr to a peak value of 11 mm=yr during a 2-ka-long period.
Discussion
The field investigations and collected data, their analysis,and results provide a remarkable succession of past earth-quakes and related rupture parameters along a single segmentof the DSF. We have identified several issues that can be sum-marized in: (1) the complex alluvial stratigraphy (hiatus andchanneling) and related uncertainties on radiocarbon datingsled to a partial identification of some paleoseismic events(e.g., A, B, C, R, and T); (2) the possibility that our paleo-seismic record is short of a few events; and (3) the accuracyin the identification of seismic clusters, episodes of quies-cence, and definition of interseismic periods. However, thecombination of paleoseismic trenching, historical studies, ar-chaeoseismic results, and their relationships allowed a mostfavorable characterization of paleoseismic events Z–V.
The analysis of paleoseismic results revealed nine fault-ing events that we attempted to correlate (Table 3) with threehistorical events (A.D. 1033, A.D. 749, and 759 B.C.) and four
archaeoseismic events (759 B.C., 1150 B.C., 2300 B.C., and2900 B.C.). The oldest part of our catalog is based on paleo-seismic trenching alone, which we strengthened by exploringany alternative interpretation. In trench T4, event VT4 is in-terpreted on the basis of increasing displacement with depthalong FZ3 and related splays, suggesting that unit i1 has beenaffected by at least one extra event with respect to unit h. Analternative explanation might be that unit h is isopach, whileunit i1 is an erosive channel fill with a noticeable dip to thewest–southwest (direction of flow). In this case, any horizon-tal movement would produce apparent vertical displacement:all offsets described along FZ3 could have occurred duringevent ZT4 only. However, our interpretation is supported by asecond observation close to the main shear zone (see labels inFig. 5d). Besides, the upper unit c appears to have beendeposited over a preexisting scarp across the main shear zone(FZ1) or to have been warped with ∼10 cm vertical displace-ment (Fig. 5d). In the former case, the most recent displace-ment would have occurred between A.D. 87 and A.D. 1920,thus indicating either of the historical events of A.D. 749 andA.D. 1033. In the latter case, warping would have taken placeafter the deposition of unit c and before the deposition of thewedge unit b (i.e., after A.D. 1660) at a time when no largeearthquake is attested for the region, which is very unlikely.
11.2 mm/yr 8.3 mm/yr
30
40
50
60
70
80
ry/mm6.1
ry/mm5.3ry/mm2.4
11.2
mm
/yr
ry/m
m3.8
0
10
20
30
40
50
60
70
80
Age (ka BP)
Ferry et al., 2007
This study
Cum
ulat
ive
slip
(m
)
0 2 4 6 8 10 12 14 16 18
4.9mm/yr
4.3 mm/yr
Figure 10. Comparison between geomorphological data (gray curve, Ferry et al., 2007) and paleoseismological data (black curve, thisstudy) for the last 14 ka. With an average coseismic displacement of 3.3 m (see text), inferred slip rate reaches 4:2 mm=yr for the last 5 ka,slightly higher than the minimum value and slightly lower than the average long-term value given by Ferry et al. (2007). Inferred cumulativeslip from historical and archaeological events (19:8� 1:8 m) satisfyingly corresponds to the first geomorphic marker (17� 5 m).Furthermore, our oldest events (O, P, and Q) suggest a short-term slip rate of ∼8:3 mm=yr, comparable to the 11:2 mm=yr value fromgeomorphology.
62 M. Ferry, M. Meghraoui, N. Abou Karaki, M. Al-Taj, and L. Khalil
Over the four trench exposures, we collected and dated 28radiocarbon samples, most of which yielded large (> 1 mg)amounts of carbon (AoC) and adequate uncertainties (20–50 yr) and appeared to be in stratigraphic order (Table 1and Fig. 5). Some samples present uncertainties that shouldbe clarified: Samples Ka-T3-S03 (T1), Ka-T3-6N (T2), L-14(T4), and Tbc-04 (T4) present low to insufficient AoC (0.2–0.8 mg) and were not used for event determination. Althoughalkali residue (AkR) fractions of samples L-02 (T4) and L-07(T4) yield low AoC, their respective carbonatic and humicacid fractions yield similar ages that confirm the quality ofthe results. However, because sample L-02 is older than anyother sample of the section by some 5 ka, including the stra-tigraphically older samples L-07 and L-06, we consider it isredeposited and chose to reject it. Sample Tbc-18 presents anextremely low AoC and needed to be normalized to speciallyprepared targets. Because it is used to define event R (paleo-seismic event VT4) and rejuvenation is very likely, we reliedon nearby samples and stratigraphy to correct its age (seedescription, Trench 4). We used the same approach for sam-ples Tbc-23 and Tbc-24, which bracket event T (paleoseis-mic event XT4) and show a relatively low AoC (0.6 mg each),similar ages, and age inversion. Although we applied com-pensation for rejuvenation of the samples, it is possible thatevents R and T are actually slightly older. This would moveevent T closer to event S and event R closer to events O, P,and Q, while extending the empty period between events Rand S.
Although we combined three independent datasets into auniquely long catalog of large earthquakes for the JVF, it is
possible that some events are still missing. The mean recur-rence interval for our paleoseismic record is 1480 yr, almostdouble the 787 yr mean recurrence interval obtained for theassumedly complete historical and archaeological periods(i.e., the last 14,000 years). This may point either to theincompleteness of the long-term paleoseismic record (oneextra event has been identified but not dated) or to differentfaulting behaviors over the two periods. Considering thesituation of alluvial deposits in channels and the relativelylong sedimentary hiatus observed in trenches (Fig. 5), it islikely that part of the sedimentary record of paleoseismicevents has been truncated. Furthermore, plotting age versuscumulative displacement for our catalog (Fig. 10) reveals thatthe paleoseismic period does not fit cumulative offsetsmeasured on streams. This may be only partly explainedby distributed off-fault deformation (Griffith et al., 2010).
Our catalog of seismic events allows us to identify earth-quake clusters and quiescence periods and suggests episodi-city for the JVF over the last 14 ka. It appears that themajority of events from our catalog (9 events out of 12)is grouped into clusters of two or three earthquakes: eventsY and Z (cluster YZ, interval of 284 yr), events W and X(WX, interval of 391 yr), events U and V (UV, interval of600 yr), and events O, P, and Q (OPQ, mean interval of330 yr). These clusters are generally preceded and followedby long periods of quiescence: up to 1800 yr after clusterOPQ, 2335 yr before cluster UV, 1150 yr between clustersUV and WX, 1508 yr between clusters WX and YZ, and atleast 977 yr after cluster YZ (time since the A.D. 1033 his-torical earthquake). Although one may also consider events
Table 3Summary of Events Identified from Historical, Archaeological, and Paleoseismic Data along the Jordan
Valley Fault for the Last 18.5 ka
Event Records Date* Interval† Confidence Reference
Z A.D. 1033 284� 1 5
ZH A.D. 1033 Ambraseys et al., 1994; Amiran et al., 1994;Ben-Menahem, 1991
ZT2 Ult: > A:D: 560–660�x1; x3; 0� This studyZT3 Ult: < A:D: 1490–1640 This studyZT4 Ult: > A:D: 500 This study
Y A.D. 749 1508� 2 5
YH A.D. 749 Abou Karaki, 1987; Tsafrir and Foerster, 1992;Karcz, 2004
YT1 Pen: < A:D: 1490–1800 This studyYT2 Pen: > A:D: 560–660 This study
X 759 B:C:� 1 391� 51 4
XH 759 B:C:� 1 Ben-Menahem, 1991; Nur and Ron, 1996;Ambraseys, 2005
ZA Mid-eighth century B.C. Tubb, 1988; Franken, 1989W YA 1150 B:C:� 50 1150� 100 3 Franken, 1989; Nur and Cline, 2000V 2300 B:C:� 50 600� 100 4
XA ∼2300 B:C: Savage et al., 20033YKI70falFSivtJa
*Ult., ultimate; Pen., penultimate.†Time interval from preceding event.
Episodic Behavior of the Jordan Valley Section of the Dead Sea Fault 63
A, B, and C as a cluster, the large uncertainty associated withradiocarbon dating (�1500 yr) implies that the three eventsmay have occurred any time within a 3-ka period.
Considering that the inland 120-km-long JVF mayextend southward inside the Dead Sea basin for a further∼20 km (Lazar and Ben-Avraham, 2002; Ben-Avraham andSchubert, 2006) and northward in the Hula asin for ∼10 km(Marco et al., 2005), this yields a maximum ∼150 km sur-face-rupture length. Hence, some of our inferences rely on amaximum Mw 7.2–7.4 magnitude and a related average∼3:3 m coseismic left-lateral slip estimated from a maximum120–150-km surface rupture length (Kanamori and Ander-son, 1975; Wells and Coppersmith, 1994). Although the22 November 1995 Mw 7.3 Aqaba event is the only largemodern earthquake observed along the DSF until today, itsrupture parameters may not serve as a base of comparisonfor earthquakes along the JVF. Indeed, it was composedof two or three subevents along offshore faults (Hofstetter,2003), and no surface rupture could be observed so far.Unfortunately, the only coseismic slip values available weremodeled and may not be adequately compared to surfacerupture values (Hofstetter, 2003). However, several coseis-mic or cumulative displacements were actually measuredalong other segments of the DSF and typically show coseis-mic values around 3 m (e.g., Klinger et al., 2000; Gomezet al., 2003; Meghraoui et al., 2003; Marco et al., 2005;Haynes et al., 2006), which suggests that a similar valuemay be considered for the JVF. Furthermore, results of recentearthquakes, such as the 1999 Mw 7.3–7.4 Izmit earthquake(Turkey) and related right-lateral strike-slip faulting, yield acomparable estimate with ∼140 km rupture length and anaverage 3 m coseismic slip (Barka et al., 2002).
Conclusions
The JVF is the main source of destructive earthquakes forthe Jordan Valley region. Its characterization has crucialimplications for the seismic hazard assessment of large urbanareas such as Jerusalem, Amman, and Irbid, as well as tonumerous historical and archaeological heritage sites suchas Pella, Jerash, Madaba, Qumran, Jericho, and Meguiddo(Fig. 6).
Through an integrated approach involving (1) the com-pilation of historical seismicity, (2) the careful reappraisal ofarchaeological data, and (3) detailed fault mapping, tectonicgeomorphology, and paleoseismic trenching, we produce anoriginal catalog of at least 12 surface-rupturing events in thelast 14 ka and at least 16 events in the last 17 ka along theJordan Valley segment of the Dead Sea fault. The meanrecurrence interval (787 yr) indicates that the historicaland archaeological record might be complete, while lowerand upper bounds of the extreme recurrence intervals(284–1508 yr) imply a generally time-episodic behavior.The plausibility of this episodic model should be comparedto apparent episodicity of conventional renewal models (Fit-zenz et al., 2010). In contrast, the paleoseismological dataset
shows incompleteness with a mean recurrence interval of1480 yr, suggesting sedimentary hiatuses and a gap in thegeological record.
Taking into account that the historical and archaeologi-cal datasets provide a mostly complete catalog of surface-rupturing events, we obtain a 787-yr mean recurrence inter-val, ∼3:3 m of slip per event (derived from Wells and Cop-persmith, 1994), and a mean 5 mm=yr slip rate for the last5 ka (or 4:8 mm=yr, using a regression curve; see Fig. 10), inagreement with the mean value obtained by Ferry et al.(2007) for the last 48 ka. This is also confirmed by the directmeasurement of a stream offset at Tell Saydiyeh that yields4:9� 0:3 mm=yr for the last 25 ka (see Tectonic Geomor-phology along the Jordan Valley Fault). Finally, the last mil-lennium of seismic quiescence along the JVF indicates up to5 m of slip deficit and points to either an imminent earth-quake and/or a future earthquake cluster similar to the A.D.749/A.D. 1033 sequence. Neighboring segments being at aslightly lower, if not similar, level of tectonic loading (i.e.,no large events since the twelfth century sequence), it is plau-sible that a present-day large earthquake on the JVF maytrigger earthquake ruptures on nearby segments.
Data and Resources
Digital elevation model data used to produce maps arefrom Shuttle Radar Topography Mission (SRTM) 3 topogra-phy from the National Aeronautics and Space Administrationand are available from Consultative Group for InternationalAgriculture Research–Consortium for Spatial InformationCGIAR–CSI Consortium for Spatial Information at srtm.csi.cgiar.org (last accessed March 2009). Bathymetry used formaps in Figure 1 and Figure 2 is SRTM 30� produced byUni-versity of California, San Diego (http://topex.ucsd.edu/). In-strumental seismicity in Figure 2 was obtained from theIncorporated Research Institutions for SeismologyDataMan-agement Center at www.iris.edu (last accessed April 2007).Calibration of radiocarbon ages was performed using theOxCal software (Bronk Ramsey, 1995), available at c14.arch.ox.ac.uk/embed.php?File=oxcal.html with the IntCal04calibration curve (Reimer et al., 2004).
Acknowledgments
The authors are indebted to the Deanship of Scientific Research (Uni-versity of Jordan), the Jordan Valley Authority, the Natural ResourcesAuthority and Military Commandment, and Salman Al-Dhaisat (Royal Jor-danian Geographic Center) for their assistance and help during our field in-vestigations. We are grateful to Majdi Barjous (Natural ResourcesAuthority), Hani Amoush (University of Jordan), and Zoe Shipton andJames Kirkpatrick (Glasgow University) for their help in the field and toPieter Grootes and Marie-Josée Nadeau (Kiel University) for the radiocar-bon dating. This study was funded by the EC 5th Framework Program andAPAME project (Contract ICA3-CT-2002-10024). Matthieu Ferry was inpart supported by the COMPETE program (FCT Ciencia 2007 FCOMP-01-0124-FEDER-009326).
64 M. Ferry, M. Meghraoui, N. Abou Karaki, M. Al-Taj, and L. Khalil
References
Abed, A. M., and R. Yaghan (2000). On the paleoclimate of Jordan duringthe last glacial maximum, Palaeogeogr. Palaeocl. 160, 23–33.
Abou Karaki, N. (1987). Synthése et carte sismotectonique des pays de labordure orientale de la Méditerrannée: Sismicité du systéme de faillesdu Jourdain—Mer Morte, Ph.D. Thesis, Université Louis Pasteur,Strasbourg.
Akyüz, H., E. Altunel, V. Karabacak, and C. Yalciner (2006). Historicalearthquake activity of the northern part of the Dead Sea fault zone,southern Turkey, Tectonophysics 426, 281–293.
Alchalbi, A., M. Daoud, F. Gomez, S. McClusky, Reilinger, R.,M. A. Romeyeh, A. Alsouod, R. Yassminh, B. Ballani, R. Darawcheh,R. Sbeinati, Y. Radwan, R. A. Masri, M. Bayerly, R. A. Ghazzi, andM. Barazangi (2010). Crustal deformation in northwestern Arabiafrom GPS measurements in Syria: Slow slip rate along the northernDead Sea fault, Geophys. J. Int. 180, 125–135.
Aldersons, F., Z. Ben-Avraham, A. Hofstetter, E. Kissling, andT. Al-Yazjeen (2003). Lower-crustal strength under the Dead Sea basinfrom local earthquake data and rheological modeling, Earth Planet.Sci. Lett. 214, 129–142.
Al-Taj, M. (2000). Active faulting along the Jordan Valley segment of theJordan—Dead Sea Transform, Ph.D. Thesis, University of Jordan,Amman.
Al-Tarazi, E., E. Sandvol, and F. Gomez (2006). The February 11, 2004Dead Sea earthquakeML � 5:2 in Jordan and its tectonic implication,Tectonophysics 422, 149–158.
Altunel, E., M. Meghraoui, V. Karabacak, S. Akyüz, M. Ferry, Ç. Yalçiner,and M. Munschy (2009). Archaeological sites (Tell and Road) offsetby the Dead Sea fault in the Amik Basin, Southern Turkey, Geophys.J. Int. 179, 1313–1329.
Ambraseys, N. N. (2004). The 12th century seismic paroxysm in the MiddleEast: A historical perspective, in Investigating the Records of PastEarthquakes, 21st course of the International School of GeophysicsBologna, Italy, 733–758.
Ambraseys, N. (2005). Historical earthquakes in Jerusalem—A methodolo-gical discussion, J. Seismol. 9, 329–340.
Ambraseys, N. N. (2006). Earthquakes and Archeology, J. Archaeol. Sci. 33,1008–1016.
Ambraseys, N. N. (2009). Earthquakes in the Mediterranean and MiddleEast, A Multidisciplinary Study of Seismicity up to 1900, CambridgeUniversity Press, Cambridge, 947 pp.
Ambraseys, N. N., and J. A. Jackson (1998). Faulting associated withhistorical and Recent earthquakes in the eastern Mediterranean region,Geophys. J. Int. 133, 390–406.
Ambraseys, N. N., C. P. Melville, and R. D. Adams (1994). The Seismicity ofEgypt, Arabia and the Red Sea: A Historical Review, CambridgeUniversity Press, Cambridge, 181 pp.
Amiran, D. H. K., E. Arieh, and T. Turcotte (1994). Earthquakes in Israel andadjacent areas: Macroseismic observations since 100 B.C.E, Isr.Explor. J. 44, 260–305.
Arieh, E., Y. Rotstein, and U. Peled (1982). The Dead Sea earthquake of 23April 1979, Bull. Seismol. Soc. Am. 72, 1627–1634.
Avni, R., D. D. Bowman, A. Shapira, and A. Nur (2002). Erroneousinterpretation of historical documents related to the epicenterof the 1927 Jericho earthquake in the Holy Land, J. Seismol. 6,469–476.
Barka, A., H. S. Akyüz, G. Sunal, Z. Çakir, A. Dikbas, B. Yerli, E. Altunel,R. Armijo, B. Meyer, J.-B. de Chabalier, T. K. Rockwell, J. R. Dolan,R. Hartleb, T. Dawson, S. Christofferson, A. Tucker, T. Fumal,R. Langridge, H. Stenner, W. Lettis, J. Bachhuber, and W. Page(2002). The surface rupture and slip distribution of the 17 August1999 Izmit earthquake (M 7.4), North Anatolian fault, Bull. Seismol.Soc. Am. 92, 43–60.
Bartov, Y., A. Agnon, Y. Enzel, and M. Stein (2006). Late Quaternaryfaulting and subsidence in the central Dead Sea basin, Isr. J. EarthSci. 55, 18–31.
Bartov, Y., M. Stein, Y. Enzel, A. Agnon, and Z. E. Reches (2002). Lakelevels and sequence stratigraphy of Lake Lisan, the Late Pleistoceneprecursor of the Dead Sea, Quat. Res. 57, 9–21.
Ben-Avraham, Z., and G. Schubert (2006). Deep “drop down” basin in thesouthern Dead Sea, Earth Planet. Sci. Lett. 251, 254–263.
Ben-Avraham, Z., R. Hanel, and H. Villinger (1978). Heat flow through theDead Sea rift, Mar. Geol. 28, 253–269.
Ben-Menahem, A. (1991). Four thousand years of seismicity along the DeadSea rift, J. Geophys. Res. 96, no. B12, 20,195–20,216.
Bronk Ramsey, C. (1995). Radiocarbon calibration and analysis ofstratigraphy: The OxCal program, Radiocarbon 37, 425–430.
Daëron, M., L. Benedetti, P. Tapponnier, A. Sursock, and R. C. Finkel(2004). Constraints on the post ~25-ka slip rate of the Yammoûnehfault (Lebanon) using in situ cosmogenic 36Cl dating of offsetlimestone-clast fans, Earth Planet. Sci. Lett. 227, 105–119.
Daëron, M., Y. Klinger, P. Tapponnier, A. Elias, E. Jacques, and A. Sursock(2007). 12,000-year-long record of 10 to 13 paleoearthquakes on theYammoûneh fault, Levant fault system, Lebanon, Bull. Seismol. Soc.Am. 97, 749–771.
El-Isa, Z. H., and H. Mustafa (1986). Earthquake deformations in the Lisandeposits and seismotectonic implications, Geophys. J. R. Astr. Soc. 86,413–424.
Ellenblum, R., S. Marco, A. Agnon, T. K. Rockwell, and A. Boas (1998).Crusader castle torn apart by earthquake at dawn, 20 May 1202,Geology 26, 303–306.
Ferry, M., and M. Meghraoui (2008). Reply to comment of Dr M. Klein on:“A 48-kyr-long slip rate history for the Jordan Valley segment of theDead Sea fault”, Earth Planet. Sci. Lett. 268, 241–242.
Ferry, M., M. Meghraoui, N. Abou Karaki, M. Al-Taj, H. Amoush,S. Al-Dhaisat, and M. Barjous (2007). A 48-kyr-long slip rate historyfor the Jordan Valley segment of the Dead Sea fault, Earth Planet. Sci.Lett. 260, 394–406.
Fitzenz, D. D., M. A. Ferry, and A. Jalobeanu (2010). Long-term slip historydiscriminates among occurrence models for seismic hazard assess-ment, Geophys. Res. Lett. 37, L20307, doi 10.1029/2010GL044071.
Franken, H. J. (1989). Deir' Alla (Tell) in Archaeology of Jordan, Vol. II1.-II2. Field Reports, Surveys and Sites, D. Homès-Fredericq andJ. Hennessy (eds), Peeters, Leuven, Belgium.
Garfunkel, Z., I. Zak, and R. Freund (1981). Active faulting in the Dead Searift, in The Dead Sea Rift: Selected Papers of the InternationalSymposium on the Dead Sea Rift Amsterdam, Netherlands, 1–26.
Ginat, H., Y. Enzel, and Y. Avni (1998). Translocated Plio-Pleistocenedrainage systems along the Arava fault of the Dead Sea Transform,Tectonophysics 284, 151–160.
Ginzburg, A., and Z. Ben-Avraham (1997). A seismic refraction studyof the north basin of the Dead Sea, Israel, Geophys. Res. Lett. 24,2063–2066.
Griffith, W. A., S. Nielsen, G. Di Toro, and S. A. F. Smith Rough faults,distributed weakening, and off-fault deformation, J. Geophys. Res.,115, no. B08409, doi 10.1029/2009JB006925.
Gomez, F., M. Meghraoui, A. Darkal, F. Hijazi, M. Mouty, Y. Sulaiman,R. Sbeinati, R. Darawcheh, R. Al-Ghazzi, and M. Barazangi (2003).Holocene faulting and earthquake recurrence along the Serghayabranch of the Dead Sea fault system in Syria and Lebanon, Geophys.J. Int. 153, 658–674.
Gomez, F., T. Nemer, C. Tabet, M. Khawlie, M. Meghraoui, andM. Barazangi (2007). Strain partitioning of active transpression withinthe Lebanese restraining bend of the Dead Sea fault (Lebanon andSW Syria), Geological Society, London, Special Publications 290,285–303.
Guidoboni, E. (1989). I terremoti prima del Mille in Italia e nell’areamediterranea, Storia-Geofisica-Ambiente, Bologna, 768.
Guidoboni, E., A. Comastri, and G. Traina (1994).Catalog of Ancient Earth-quakes in the Mediterranean Area up to the 10th Century, ING-SGA,Bologna, 504 pp.
Haynes, J., T. Niemi, and M. Atallah (2006). Evidence for ground-rupturingearthquakes on the Northern Wadi Araba fault at the archaeological
Episodic Behavior of the Jordan Valley Section of the Dead Sea Fault 65
site of Qasr Tilah, Dead Sea Transform fault system, Jordan, J. Seis-mol. 10, 415–430.
Hofstetter, A. (2003). Seismic observations of the 22/11/1995 Gulf of Aqabaearthquake sequence, Tectonophysics 369, 21–36.
Jordan Seismological Observatory (2004). The Dead Sea earthquake,ML = 4.9—a preliminary report, Natural Resources Authority, TheHashemite Kingdom of Jordan, 6 pp.
Kanamori, H., and D. G. Anderson (1975). Theoretical basis of someempirical relations in seismology, Bull. Seismol. Soc. Am. 65,1073–1095.
Karabacak, V., E. Altunel, M. Meghraoui, and H. S. Akyüz (2010). Fieldevidences from northern Dead Sea fault zone (south Turkey): Newfindings for the initiation age and slip rate, Tectonophysics 480,172–182.
Karcz, I. (2004). Implications of some early Jewish sources for estimatesof earthquake hazard in the Holy Land, in Investigating the Recordsof Past Earthquakes, 21st course of the International School ofGeophysics Bologna, Italy, 759–792.
Katz, O., and O. Crouvi (2007). The geotechnical effects of long humanhabitation (2000 < years): Earthquake induced landslide hazard inthe city of Zefat, northern Israel, Eng. Geol. 95, 57–78.
Ken-Tor, R., A. Agnon, Y. Enzel, M. Stein, S. Marco, and J. Negendank(2001). High-resolution geological record of historic earthquakes inthe Dead Sea basin, J. Geophys. Res. 106, 2221–2234.
King, G., G. Bailey, and D. Sturdy (1994). Active tectonics and humansurvival strategies, J. Geophys. Res. 99, 20,063–20,078.
Klein, M. (2008). A comment on: “A 48-kyr-long slip rate history forthe Jordan Valley segment of the Dead Sea fault” EPSL 260(2007). 394–406, Earth Planet. Sci. Lett. 268, 239–240.
Klinger, Y., J.-P. Avouac, N. Abou Karaki, L. Dorbath, D. Bourles, andJ. L. Reyss (2000). Slip rate on the Dead Sea Transform in northernAraba Valley (Jordan), Geophys. J. Int. 142, 755–768.
Lazar, M., and Z. Ben-Avraham (2002). First images from the bottom of theDead Sea—Indications of recent tectonic activity, in Honoring EytanSass and Amitai Katz on the occasion of their 70th birthdays, Jerusa-lem, Israel, 211–218.
Le Beon, M., Y. Klinger, A. Q. Amrat, A. Agnon, L. Dorbath, G. Baer,J.-C. Ruegg, O. Charade, and O. Mayyas (2008). Slip rate and lockingdepth from GPS profiles across the southern Dead Sea Transform,J. Geophys. Res. 113, no. B11403, doi 10.1029/2007JB005280.
Malkawi, A. I. H., and A. S. Alawneh (2000). Paleoearthquake features asindicators of potential earthquake activities in the Karameh Dam site,Nat. Hazards 22, 1–16.
Marco, S., M. Hartal, N. Hazan, L. Lev, and M. Stein (2003). Archaeology,history, and geology of the A.D. 749 earthquake, Dead Sea Transform,Geology 31, 665–668.
Marco, S., T. K. Rockwell, A. Heimann, U. Frieslander, and A. Agnon(2005). Late Holocene activity of the Dead Sea Transform revealedin 3D paleoseismic trenches on the Jordan Gorge segment, EarthPlanet. Sci. Lett. 234, 189–205.
Marco, S., M. Stein, A. Agnon, and H. Ron (1996). Long-term earthquakeclustering: A 50,000-year paleoseismic record in the Dead Sea graben,J. Geophys. Res. 101, 6179–6192.
McCalpin, J. P. (1998). Paleoseismology, Academic Press, New York,588 pp.
McClusky, S., R. Reilinger, S. Mahmoud, S. D. Ben, and A. Tealeb (2003).GPS constraints on Africa (Nubia) and Arabia plate motions,Geophys.J. Int. 155, 126–138.
Meghraoui, M., F. Gomez, R. Sbeinati, J. Van der Woerd, M. Mouty,A. Darkal, Y. Radwan, I. Layyous, H. M. Najjar, R. Darawcheh,F. Hijazi, R. Al-Ghazzi, and M. Barazangi (2003). Evidence for830 years of seismic quiescence from paleoseismology, archeoseis-mology and historical seismicity along the Dead Sea fault in Syria,Earth Planet. Sci. Lett. 210, 35–52.
Meyers, E., A. Kraabel, and J. Strange (1976). Ancient synagogue excava-tions at Khirbet Shema’, Upper Galilee, Israel 1970–1972, Annu. Am.Sch. Orient. Res. 42.
Migowski, C., A. Agnon, R. Bookman, J. F. W. Negendank, and M. Stein(2004). Recurrence pattern of Holocene earthquakes along the DeadSea Transform revealed by varve-counting and radiocarbon datingof lacustrine sediments, Earth Planet. Sci. Lett. 222, 301–314.
Nemer, T., and M. Meghraoui (2006). Evidence of coseismic ruptures alongthe Roum fault (Lebanon): A possible source for the AD 1837 earth-quake, J. Struct. Geol. 28, 1483–1495.
Nemer, T., M. Meghraoui, and K. Khair (2008). The Rachaya-Serghaya faultsystem (Lebanon): Evidence of coseismic ruptures, and the AD 1759earthquake sequence, J. Geophys. Res. 113, no. B05312, doi 10.1029/2007JB005090.
Niemi, T. M., and N. Mansoor (2002). Nearly a millennium of seismic quies-cence in Aquaba, Jordan along the southern Dead Sea Transform,paper no. 227, Annual Meeting of the Geological Society of America,Denver, Colorado, 27–30 October 2002, 227–210.
Niemi, T. M., H. Zhang, M. Atallah, and J. B. J. Harrison (2001). LatePleistocene and Holocene slip rate of the northern Wadi Araba fault,Dead Sea Transform, Jordan, J. Seismol. 5, 449–474.
Nur, A., and E. H. Cline (2000). Poseidon’s Horses: Plate tectonics andearthquake storms in the Late Bronze Age Aegean and easternMediterranean, J. Archaeol. Sci. 27, 43–63.
Nur, A., and H. Ron (1996). And the walls came tumbling down: Earthquakehistory in the Holy Land, in Archaeoseismology, S. Stiros and R. Jones(eds), British School at Athens, Athens, Greece, 75–85.
Petrunin, A., and S. V. Sobolev (2006). What controls thickness of sedimentsand lithospheric deformation at a pull-apart basin?, Geology 34,389–392.
Poirier, J.-P., and M. A. Taher (1980). Historical seismicity in the near andMiddle East, North Africa, and Spain from Arabic documents (VIIth-XVIIIth century), Bull. Seismol. Soc. Am. 70, 2185–2201.
Quennell, A. M. (1959). Tectonics of the Dead Sea rift, in 20thInternational Geological Congress, Association of African GeologicalSurveys, 385–405.
Quennell, A. M (1984). The Western Arabia Rift System, Geological Society,London, Special Publications 17, 775–788.
Reches, Z., and D. F. Hoexter (1981). Holocene seismic and tectonic activityin the Dead Sea area, in The Dead Sea Rift: Selected Papers of theInternational Symposium on the Dead Sea Rift Amsterdam,Netherlands, 235–254.
Reilinger, R., S. McClusky, P. Vernant, and S. Lawrence (2006).GPS constraints on continental deformation in the Africa-Arabia-Eurasia continental collision zone and implications for the dynamicsof plate interactions, J. Geophys. Res. 111, doi 10.1029/2005JB004051.
Reimer, P. J.,M.G.L.Baillie, E.Bard,A.Bayliss, J.W.Beck, P.G.Blackwell,C. E. Buck, G. S. Burr, K. B. Cutler, P. E. Damon, R. L. Edwards,R.G. Fairbanks,M. Friedrich, T. P. Guilderson,C.Herring,K.A.Hugh-en, B. Kromer, F. G. McCormac, S. W. Manning, C. B. Ramsey,R. W. Reimer, S. Remmele, J. R. Southon, M. Stuiver, S. Talamo,F.W. Taylor, J. van der Plicht, and C. E.Weyhenmeyer (2004). IntCal04Terrestrial radiocarbon age calibration, 0–26 cal kyr BP, Radiocarbon43, 1029–1058.
Ryberg, T., M. Weber, Z. Garfunkel, and Y. Bartov (2007). The shallowvelocity structure across the Dead Sea Transform fault, Arava Valley,from seismic data, J. Geophys. Res. 112, no. B0830, doi 10.1029/2006JB004563.
Salamon, A., A. Hofstetter, Z. Garfunkel, and H. Ron (2003). Seismotec-tonics of the Sinai subplate—The eastern Mediterranean region,Geophys. J. Int. 155, 149-–173.
Savage, S., K. Zamora, and D. Keller (2001). Archaeology in Jordan, Am. J.Archaeol. 105, 427–461.
Savage, S., K. Zamora, and D. Keller (2002). Archaeology in Jordan, 2001Season, Am. J. Archaeol. 106, 435–458.
Savage, S., K. Zamora, and D. Keller (2003). Archaeology in Jordan, 2002Season, Am. J. Archaeol. 107, 449–475.
Savage, S., K. Zamora, and D. Keller (2005). Archaeology in Jordan, 2004Season, Am. J. Archaeol. 109, 527–555.
66 M. Ferry, M. Meghraoui, N. Abou Karaki, M. Al-Taj, and L. Khalil
Sbeinati, M. R., R. Darawcheh, and M. Mouty (2005). The historical earth-quakes of Syria: An analysis of large and moderate earthquakes from1365 B.C. to 1900 A.D, Ann. Geophys. 48, 347–435.
Sbeinati, M. R., M. Meghraoui, G. Suleyman, P. Grootes, M.-J. Nadeau,H. Al Najjar, F. Gomez, and R. Al-Ghazzi (2010). Timing of earth-quake ruptures at the Al Harif Roman aqueduct (Dead Sea fault, Syria)from archaeoseismology and palaeoseismology, special paper 471 inAncient Earthquakes (special volume), M. Sintubin, I. Stewart,T. Niemi, and E. Altunel (Editors), Geological Society of AmericaBulletin, 243–267.
Shapira, A., R. Avni, and A. Nur (1993). A new estimate for the epicenter ofthe Jericho earthquake of 11 July 1927, Isr. J. Earth Sci. 42, 93–96.
Sieh, K. (1996). The repetition of large-earthquake ruptures, Proc. Natl.Acad. Sci. Unit. States Am. 93, 3764–3771.
Stein, R. S., A. A. Barka, and J. H. Dieterich (1997). Progressive failure onthe North Anatolian Fault since 1939 by earthquake stress triggering,Geophys. J. Int. 128, 594–604.
ten Brink, U. S., Z. Ben-Avraham, R. E. Bell, M. Hassouneh, D. F. Coleman,G. Andreasen, G. Tibor, and B. J. Coakley (1993). Structure of theDead Sea pull-apart basin from gravity analyses, J. Geophys. Res.98, 21,877–21,894.
Tsafrir, Y., and G. Foerster (1992). The dating of the ’Earthquake of theSabbatical Year’ of 749 C.E. in Palestine, Bull. Sch. Orient. Afr. Stud.55, 231–235.
Tubb, J. N. (1988). Tell es-Sa’idiyeh: Preliminary report on the first threeseasons of renewed excavations, Levant 20, 23–89.
Tubb, J. N. (1998). Canaanites, The British Museum Press, London, 160.Wdowinski, S., Y. Bock, G. Baer, L. Prawirodirdjo, N. Bechor, S. Naaman,
R. Knafo, Y. Forrai, and Y. Melzer (2004). GPS measurements ofcurrent crustal movements along the Dead Sea fault, J. Geophys.Res. 109, no. B05403, doi 10.1029/2003JB002640.
Wells, D. L., and K. J. Coppersmith (1994). New empirical relationshipsamong magnitude, rupture length, rupture width, rupture area, andsurface displacement, Bull. Seismol. Soc. Am. 84, 974–1002.
Willis, B. (1928). Earthquakes in the Holy Land, Bull. Seismol. Soc. Am. 18,73–103.
Zilberman, E., R. Amit, N. Porat, Y. Enzel, and U. Avner (2005). Surfaceruptures induced by the devastating 1068 AD earthquake in the south-ern Arava valley, Dead Sea rift, Israel, Tectonophysics 408, 79–99.
Centro de Geofísica de ÉvoraUniversidade de ÉvoraRua Romão Ramalho, 597002-554 Évora, [email protected]
(M.F.)
Institut de Physique du GlobeUMR 7516, 5 rue René Descartes67084 Strasbourg, France
(M.M.)
Department of Environmental and Applied GeologyUniversity of JordanAmman 11942, Jordan
(N.A.)
Department of Earth and Environmental SciencesThe Hashemite UniversityP.O. Box 150459Zarqa 13115, Jordan
(M.A.)
Department of ArcheologyUniversity of JordanAmman 11942, Jordan
(L.K.)
Manuscript received 13 April 2010
Episodic Behavior of the Jordan Valley Section of the Dead Sea Fault 67